ORIGINAL CONTRIBUTION violence, prevention; weapons
Weapons Possession by Patients in a University Emergency Department Study objective: Violence in the emergency department, a not uzlcommon but complex phenomenon, m a y become more serious when patients possess weapons. Searches are used frequently to reduce this danger, though guidelines for searches are not well delineated. We examined our practices in order to formalize our guidelines. Design: Retrospective chart review of patients found to be carrying weapons. Setting: General, university-based emergency department in the Northwest. Participants: Of 39,000 patients seen during the 20-month study period, 500 (1.3%) were searched. Measures and main results: Of all patients seen in the ED, 92% were medical patienzs (153, 0.4% of whom were searched) and 8% were psychiatric patients (347, 11.1% of whom were searched). Weapons were found orl 89 patients (0.2% of all ED patients and 17.8% of all patients searched). Review showed that 24 (15.7%) medical and 60 (17.3%) psychiatric patients carried weapons. Conclusion: Although various factors contributed to a clear bias toward searching psychiatric patients, we believe that the rate of weapons possession did not support this bias. [Goetz RR, Bloom JD, Cheqell SL, Moorhead JC: Weapons possession by patients in a university emergency department. Ann Emerg Med Jal~uary 1991;20:8-I0.]
Rupert R Goetz, MD Joseph D Bloom, MD Sherry L Chenell, RN, MS, PMHNP John C Moorhead, MD, MS, FACEP Portland, Oregon From the Psychiatric Emergency Services, Oregon Health Sciences University, Portland. Received for publication April 2, 1990. Revision received July 5, 1990. Accepted for publication July 25, 1990. Address for reprints: Rupert R Goetz, MD, Department of Psychiatry, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, Oregon 97201.
INTRODUCTION Violence in the emergency department is not uncommon. A study of general EDs in US teaching hospitals found that 41 of 127 responding institutions (32%) reported more than one verbal threat of violence per day and that 55 (43%) reported one or more attacks on staff per month. 1 Eighteen percent of the 127 responding institutions reported that a weapon was displayed an average of once or more per month and that 46% reported confiscating a weapon at least once a month. In response to such violence, a broad series of measures have been proposed, ranging from the presence of securify personnel, alarm systems, and metal-detection capabilities to specific staff training and operating procedures. 2 In the emergency setting, patients are seen in varying degrees of physical and emotional distress. They generally present with little previous warning, and the patient and physician often have no prior relationship. This setting is closely intertwined with the surrounding community. Incidence of violent crime in communities increased by 130% between 1960 and 1984.3 Because the availability of a weapon to a possibly impulsiye patient clearly increases the potential for injury, it is not surprising that attention has turned to weapons possession. Our study was prompted by a dangerous incident in which a gun was pointed at a security guard who was standing next to one of the authors and represents an attempt to examine procedures in the area of weapons possession, not only for review of our institution's practice but also to contribute to this discussion. METHODS The study was a retrospective chart review of patients found to be carry-
20:1 January 1991
Annals of Emergency Medicine
8/27
WEAPONS Goetz et al
FIGURE. Study design: Patients seen per group.
ing weapons in this institution's ED. Oregon Health Sciences University, Portland, Oregon, is a university-based ED with one central entrance. Uniformed, hospital-trained security officers are present at this location on a 24-hour-a-day basis. The department is staffed by emergency physicians. A distinct psychiatric emergency service is physically integrated in the ED setting and staffed by psychiatric residents and psychiatric mental health nurse practitioners. Patients entering the ED can be triaged directly to emergency medicine staff or to emergency psychiatric staff. Consults on patients triaged to one service were obtained freely in both directions, and patients could be admitted to either psychiatry or other services. Data on patients admitted to the ED are entered into the hospital administrative computer. A separate psychiatric log is kept on all patients seen by the psychiatric emergency service, and patient contacts by security are logged into the security department's computer. At the time of study, no specific, written search policy was in effect. Patients and/or visitors identified as possibly dangerous by either the security officers or medical personnel could be given the option of being searched or leaving the ED. Although not written into policy, it was the recommendation of the director of the psychiatric emergency service and the University Department of Public Safety to have a low index of suspicion when requesting searches. With its administrative separation, a significant bias toward searching psychiatric patients preferentially was the practical outcome. From January 1, 1987, through August 31, 1988, security officers kept a list of the 500 persons searched in the ED; these searches produced a total of 89 potential weapons. The charts for these cases (a total of 72 separate patient records) were reviewed. RESULTS For the purposes of this study, patients triaged directly to the psychiatry service or seen in consultation by the psychiatric emergency service after initial evaluation were considered " p s y c h i a t r i c p a t i e n t s . " Recorded 28/9
ED ~- Patients \ 39,005 \ (100.0%) ~ All
Medical 35,874 ~ (92.0%)
Psychiatric ~-X-~ 3,131 (8.0%)
onto the psychiatric emergency service log were 3,131 patients during the study interval, which represented 8% of the total ED 39,005 patients. Thus, 92% of all ED patients (35,874) represented "medical patients." We next considered the security officer's list of 500 searched patients. Given the 39,005 patients seen, we calculated that 1.3% of all ED patients were searched. Among the 500 patients, 89 potential weapons were found (17.8%). Seventy-seven of the w e a p o n s w e r e k n i v e s ; 11 w e r e w e a p o n s s u c h as r a z o r b l a d e s , truncheons, and nunchucks, and one was a firearm (pellet gun). C h a r t s were r e v i e w e d for the w e a p o n s - c a r r y i n g p a t i e n t s (89 weapons representing 72 individuals). Because the information for which the charts were reviewed was not always recorded in each chart, different numbers resulted for some questions. Demographic i n f o r m a t i o n for the searched patients showed them to be predominantly male with an average age of 34 years, similar to the weapons-carrying patients (Table). In the ED, when such observations were r e c o r d e d , t h e b e h a v i o r of t h e weapons-carrying patients was noted to be loud and/or belligerent in 17 of the cases (19.1%), m a k i n g verbal threats in three cases (3.*4%), and physically dyscontrolled in two cases (2.2%). Five patients (5.9%) required restraints because of dyscontrol. The patients had an average of 7.3 prior ED visits. The breakdown by medical versus psychiatric patients showed that 0.4% of all medical (153) and 11.1% of all psychiatric (347) patients seen in the ED during the study interval Annals of Emergency Medicine
Not Searched 35,721 / (99.6%)
No Weapon (84.3%)
Searched 153 (0.4%)
Weapon 24 (15.7%)
Not Searched 2,784 / (88.9%°)
287
Searched 347 (11.1%)
Weapon 60 (17.3%)
129
No Weapon (82.7%)
TABLE. Demographic information
Patients Searched (%)
Male Female Averageage (yr) Single Married Nonveterans Veterans
363 (72,6) 137 (27,4) 33,9
Patients Found to Be Carrying Weapons (%)
74 (83.1) 15 (16.9) 33.6 66 (74.2) 34 (25.8) 73 (69.2) 16 (30.8)
Total number of patients searched was 500.
were searched. Weapons were found in 15.7% of the medical (24) and 17.3% of the psychiatric (60) patients searched (Figure). Further comparison of the medical and psychiatric groups showed that the average n u m b e r of prior ED visits was similar (9.4 for medical and 7.0 for psychiatric patients). The p r i m a r y clinical diagnoses among the medical weapons-carrying patients were extremely varied, ranging f r o m p o l y t r a u m a and s t a b wounds to rash and chronic low back pain. Eighteen psychiatric patients were also given secondary physical d i s o r d e r or c o n d i t i o n dzagnoses, whereas five "medical" patients also received secondary psychiatric clinical syndrome diagnoses. The most frequent primary diagnoses among the 60 psychiatric weapons-carrying patients were substance abuse (36.7%) and p s y c h o t i c disorders (31.7%). One patient was given a primary diagnosis of a personality disorder in the ED setting.
20:1 January1991
WEAPONS Goetz et al
DISCUSSION Our study had several limitations. The study was retrospective in nature. Search criteria were not clearly delineated, and bias other than the clearly noted emphasis on searching patients seen through the psychiatric emergency service was difficult to delineate. Demographic information may be skewed as the security officers and clinicians believed that they were apt to search young male patients preferentially. Our definition of a "weapon" was arbitrary, because a folding knife can represent a harmless tool, yet other implements such as combs or pencils can turn into dangerous weapons. Nonsuspicion-based criteria, however, give rise to Fourth Amendment concerns of arbitrary or capricious search. A truly random study may be legally problematic. In our study, the estimated percentage of patients found to be carrying weapons was comparable to that described by others.¢ s It is clear that p s y c h i a t r i c p a t i e n t s w e r e searched in much greater proportion than medical patients; medical patients constituted 69.4% of our sample, and psychiatric patients represented only 8.0% of the patients seen in our ED. It appears equally clear that psychiatric patients were no more likely to be carrying a weapon than were medical patients. Given our administrative structure, we feel
20:1 January 1991
quite confident that such a comparison of our medical with our psychiatric p a t i e n t s can be c o n s i d e r e d valid. It remains to be studied whether weapons possession actually defines a more dangerous group of patients in terms of outcome (eg, patient or staff injury). Attempts to compare our findings with those of others emphasize the importance of standardizing definitions of terms used and data gathered. Institutions have begun to do this in a variety of ways. For example, D r u m m o n d and Sparr reported the use of a behavioral emergency c o m m i t t e e to track incidents, train staff, and formulate policy.6, 7 The task, namely, to increase the safety of our EDs for both patients and staff, is clear. Issues of identification, including the decision to search patients, must be clarified.a, 9 Management (including security interventions, medical treatments, and staff training) must be an ongoing effort that includes information gathering and quality control. CONCLUSION We looked at search results to determine if our current policies were supported by these data. We found that, in terms of weapons identified, no distinction between medical and psychiatric patients was indicated. Additional research is planned to de-
Annals of Emergency Medicine
lineate the impact of this observation on the operation of our ED while developing a scientifically and legally sound basis for further study of violence in the ED setting. The authors thank the University Department of Public Safety for their efforts in compiling documentation of the searches, which made this study possible. Their consistent presence and interest provided both a secure environment in the ED and encouragement to carefully consider our procedures.
REFERENCES i. Lavoie F, Carter G, Danzl D, et al: Emergency department violence in United States teaching hospitals. Ann Emery Med 1988~17:IZ~7-124g. 2. American College of Emergency Physicians: Emergency Department VioIence: Prevention and Management. Dallas, ACEP, 1988.
3. Sourcebook of Criminal Justice Statistics, 1985. Washington, DC, US Department of Justice, Bureau of Justice Statistics, NCJ-100899, 1986. \ 4. Anderson A, Ghali A, Bansil R: Weapon carrying among patients in a psychiatric emergency room. Hosp Community Psychiatry 1989;40:845-847. 5. McNiel D, Binder R: Patients who bring weapons to the psychiatric emergency room. J CIin Psychiatry 1987~48:230-233. 6. Drummond D, Sparr L, Gordon J: Hospital violence reduction among high risk patients. JAMA 1989;261: 25362534. 7. Sparr L, Drummond D, Hamilton G: Management of violent patient incidents: The role of a behavioral emergency committee. Q Rev Bull 1988;i4:147-153. 8. McCulloch L, McNiel D, Binder R, et ah Effects of a weapons screening procedure in a psychiatric emergency room. Hosp Community Psychiatry 1986;37: 837-838. 9. Privitera M, Springer M, Perlmutter R: To search or not to search: Is there a clinical profile of a patient harboring a weapon? Gen Hosp Psychiatry 1986;8:442-447.
10/29
Weapons Possession by Patients in a University Emergency Department Study objective: Violence in the emergency department, a not uzlcommon but complex phenomenon, m a y become more serious when patients possess weapons. Searches are used frequently to reduce this danger, though guidelines for searches are not well delineated. We examined our practices in order to formalize our guidelines. Design: Retrospective chart review of patients found to be carrying weapons. Setting: General, university-based emergency department in the Northwest. Participants: Of 39,000 patients seen during the 20-month study period, 500 (1.3%) were searched. Measures and main results: Of all patients seen in the ED, 92% were medical patienzs (153, 0.4% of whom were searched) and 8% were psychiatric patients (347, 11.1% of whom were searched). Weapons were found orl 89 patients (0.2% of all ED patients and 17.8% of all patients searched). Review showed that 24 (15.7%) medical and 60 (17.3%) psychiatric patients carried weapons. Conclusion: Although various factors contributed to a clear bias toward searching psychiatric patients, we believe that the rate of weapons possession did not support this bias. [Goetz RR, Bloom JD, Cheqell SL, Moorhead JC: Weapons possession by patients in a university emergency department. Ann Emerg Med Jal~uary 1991;20:8-I0.]
Rupert R Goetz, MD Joseph D Bloom, MD Sherry L Chenell, RN, MS, PMHNP John C Moorhead, MD, MS, FACEP Portland, Oregon From the Psychiatric Emergency Services, Oregon Health Sciences University, Portland. Received for publication April 2, 1990. Revision received July 5, 1990. Accepted for publication July 25, 1990. Address for reprints: Rupert R Goetz, MD, Department of Psychiatry, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, Oregon 97201.
INTRODUCTION Violence in the emergency department is not uncommon. A study of general EDs in US teaching hospitals found that 41 of 127 responding institutions (32%) reported more than one verbal threat of violence per day and that 55 (43%) reported one or more attacks on staff per month. 1 Eighteen percent of the 127 responding institutions reported that a weapon was displayed an average of once or more per month and that 46% reported confiscating a weapon at least once a month. In response to such violence, a broad series of measures have been proposed, ranging from the presence of securify personnel, alarm systems, and metal-detection capabilities to specific staff training and operating procedures. 2 In the emergency setting, patients are seen in varying degrees of physical and emotional distress. They generally present with little previous warning, and the patient and physician often have no prior relationship. This setting is closely intertwined with the surrounding community. Incidence of violent crime in communities increased by 130% between 1960 and 1984.3 Because the availability of a weapon to a possibly impulsiye patient clearly increases the potential for injury, it is not surprising that attention has turned to weapons possession. Our study was prompted by a dangerous incident in which a gun was pointed at a security guard who was standing next to one of the authors and represents an attempt to examine procedures in the area of weapons possession, not only for review of our institution's practice but also to contribute to this discussion. METHODS The study was a retrospective chart review of patients found to be carry-
20:1 January 1991
Annals of Emergency Medicine
8/27
WEAPONS Goetz et al
FIGURE. Study design: Patients seen per group.
ing weapons in this institution's ED. Oregon Health Sciences University, Portland, Oregon, is a university-based ED with one central entrance. Uniformed, hospital-trained security officers are present at this location on a 24-hour-a-day basis. The department is staffed by emergency physicians. A distinct psychiatric emergency service is physically integrated in the ED setting and staffed by psychiatric residents and psychiatric mental health nurse practitioners. Patients entering the ED can be triaged directly to emergency medicine staff or to emergency psychiatric staff. Consults on patients triaged to one service were obtained freely in both directions, and patients could be admitted to either psychiatry or other services. Data on patients admitted to the ED are entered into the hospital administrative computer. A separate psychiatric log is kept on all patients seen by the psychiatric emergency service, and patient contacts by security are logged into the security department's computer. At the time of study, no specific, written search policy was in effect. Patients and/or visitors identified as possibly dangerous by either the security officers or medical personnel could be given the option of being searched or leaving the ED. Although not written into policy, it was the recommendation of the director of the psychiatric emergency service and the University Department of Public Safety to have a low index of suspicion when requesting searches. With its administrative separation, a significant bias toward searching psychiatric patients preferentially was the practical outcome. From January 1, 1987, through August 31, 1988, security officers kept a list of the 500 persons searched in the ED; these searches produced a total of 89 potential weapons. The charts for these cases (a total of 72 separate patient records) were reviewed. RESULTS For the purposes of this study, patients triaged directly to the psychiatry service or seen in consultation by the psychiatric emergency service after initial evaluation were considered " p s y c h i a t r i c p a t i e n t s . " Recorded 28/9
ED ~- Patients \ 39,005 \ (100.0%) ~ All
Medical 35,874 ~ (92.0%)
Psychiatric ~-X-~ 3,131 (8.0%)
onto the psychiatric emergency service log were 3,131 patients during the study interval, which represented 8% of the total ED 39,005 patients. Thus, 92% of all ED patients (35,874) represented "medical patients." We next considered the security officer's list of 500 searched patients. Given the 39,005 patients seen, we calculated that 1.3% of all ED patients were searched. Among the 500 patients, 89 potential weapons were found (17.8%). Seventy-seven of the w e a p o n s w e r e k n i v e s ; 11 w e r e w e a p o n s s u c h as r a z o r b l a d e s , truncheons, and nunchucks, and one was a firearm (pellet gun). C h a r t s were r e v i e w e d for the w e a p o n s - c a r r y i n g p a t i e n t s (89 weapons representing 72 individuals). Because the information for which the charts were reviewed was not always recorded in each chart, different numbers resulted for some questions. Demographic i n f o r m a t i o n for the searched patients showed them to be predominantly male with an average age of 34 years, similar to the weapons-carrying patients (Table). In the ED, when such observations were r e c o r d e d , t h e b e h a v i o r of t h e weapons-carrying patients was noted to be loud and/or belligerent in 17 of the cases (19.1%), m a k i n g verbal threats in three cases (3.*4%), and physically dyscontrolled in two cases (2.2%). Five patients (5.9%) required restraints because of dyscontrol. The patients had an average of 7.3 prior ED visits. The breakdown by medical versus psychiatric patients showed that 0.4% of all medical (153) and 11.1% of all psychiatric (347) patients seen in the ED during the study interval Annals of Emergency Medicine
Not Searched 35,721 / (99.6%)
No Weapon (84.3%)
Searched 153 (0.4%)
Weapon 24 (15.7%)
Not Searched 2,784 / (88.9%°)
287
Searched 347 (11.1%)
Weapon 60 (17.3%)
129
No Weapon (82.7%)
TABLE. Demographic information
Patients Searched (%)
Male Female Averageage (yr) Single Married Nonveterans Veterans
363 (72,6) 137 (27,4) 33,9
Patients Found to Be Carrying Weapons (%)
74 (83.1) 15 (16.9) 33.6 66 (74.2) 34 (25.8) 73 (69.2) 16 (30.8)
Total number of patients searched was 500.
were searched. Weapons were found in 15.7% of the medical (24) and 17.3% of the psychiatric (60) patients searched (Figure). Further comparison of the medical and psychiatric groups showed that the average n u m b e r of prior ED visits was similar (9.4 for medical and 7.0 for psychiatric patients). The p r i m a r y clinical diagnoses among the medical weapons-carrying patients were extremely varied, ranging f r o m p o l y t r a u m a and s t a b wounds to rash and chronic low back pain. Eighteen psychiatric patients were also given secondary physical d i s o r d e r or c o n d i t i o n dzagnoses, whereas five "medical" patients also received secondary psychiatric clinical syndrome diagnoses. The most frequent primary diagnoses among the 60 psychiatric weapons-carrying patients were substance abuse (36.7%) and p s y c h o t i c disorders (31.7%). One patient was given a primary diagnosis of a personality disorder in the ED setting.
20:1 January1991
WEAPONS Goetz et al
DISCUSSION Our study had several limitations. The study was retrospective in nature. Search criteria were not clearly delineated, and bias other than the clearly noted emphasis on searching patients seen through the psychiatric emergency service was difficult to delineate. Demographic information may be skewed as the security officers and clinicians believed that they were apt to search young male patients preferentially. Our definition of a "weapon" was arbitrary, because a folding knife can represent a harmless tool, yet other implements such as combs or pencils can turn into dangerous weapons. Nonsuspicion-based criteria, however, give rise to Fourth Amendment concerns of arbitrary or capricious search. A truly random study may be legally problematic. In our study, the estimated percentage of patients found to be carrying weapons was comparable to that described by others.¢ s It is clear that p s y c h i a t r i c p a t i e n t s w e r e searched in much greater proportion than medical patients; medical patients constituted 69.4% of our sample, and psychiatric patients represented only 8.0% of the patients seen in our ED. It appears equally clear that psychiatric patients were no more likely to be carrying a weapon than were medical patients. Given our administrative structure, we feel
20:1 January 1991
quite confident that such a comparison of our medical with our psychiatric p a t i e n t s can be c o n s i d e r e d valid. It remains to be studied whether weapons possession actually defines a more dangerous group of patients in terms of outcome (eg, patient or staff injury). Attempts to compare our findings with those of others emphasize the importance of standardizing definitions of terms used and data gathered. Institutions have begun to do this in a variety of ways. For example, D r u m m o n d and Sparr reported the use of a behavioral emergency c o m m i t t e e to track incidents, train staff, and formulate policy.6, 7 The task, namely, to increase the safety of our EDs for both patients and staff, is clear. Issues of identification, including the decision to search patients, must be clarified.a, 9 Management (including security interventions, medical treatments, and staff training) must be an ongoing effort that includes information gathering and quality control. CONCLUSION We looked at search results to determine if our current policies were supported by these data. We found that, in terms of weapons identified, no distinction between medical and psychiatric patients was indicated. Additional research is planned to de-
Annals of Emergency Medicine
lineate the impact of this observation on the operation of our ED while developing a scientifically and legally sound basis for further study of violence in the ED setting. The authors thank the University Department of Public Safety for their efforts in compiling documentation of the searches, which made this study possible. Their consistent presence and interest provided both a secure environment in the ED and encouragement to carefully consider our procedures.
REFERENCES i. Lavoie F, Carter G, Danzl D, et al: Emergency department violence in United States teaching hospitals. Ann Emery Med 1988~17:IZ~7-124g. 2. American College of Emergency Physicians: Emergency Department VioIence: Prevention and Management. Dallas, ACEP, 1988.
3. Sourcebook of Criminal Justice Statistics, 1985. Washington, DC, US Department of Justice, Bureau of Justice Statistics, NCJ-100899, 1986. \ 4. Anderson A, Ghali A, Bansil R: Weapon carrying among patients in a psychiatric emergency room. Hosp Community Psychiatry 1989;40:845-847. 5. McNiel D, Binder R: Patients who bring weapons to the psychiatric emergency room. J CIin Psychiatry 1987~48:230-233. 6. Drummond D, Sparr L, Gordon J: Hospital violence reduction among high risk patients. JAMA 1989;261: 25362534. 7. Sparr L, Drummond D, Hamilton G: Management of violent patient incidents: The role of a behavioral emergency committee. Q Rev Bull 1988;i4:147-153. 8. McCulloch L, McNiel D, Binder R, et ah Effects of a weapons screening procedure in a psychiatric emergency room. Hosp Community Psychiatry 1986;37: 837-838. 9. Privitera M, Springer M, Perlmutter R: To search or not to search: Is there a clinical profile of a patient harboring a weapon? Gen Hosp Psychiatry 1986;8:442-447.
10/29
