- Email: [email protected]
72
Research Forum Abstracts Conclusion: Percutaneous mechanical embolectomy for AIS is feasible, and appears to confer both survival and functional outcome advantage compared to no intervention.
69
Recurring Adolescent Victims of Firearms-Related Injuries: A Population at Risk
Cox JA, Davidson DJ, Smock WS/University of Louisville, School of Medicine, Louisville, KY
Introduction: Adolescents that sustain a firearms-related injury are at increased risk of being repeat victims of gun-violence. Nearly 5% of adolescent firearms-related victims presenting to an urban trauma center will sustain additional firearms-related injuries. Study Objective: This study examined the characteristics of repeat gunshot wound victims that were adolescents at time of first injury. Methods: Data was reviewed from the University of Louisville Gunshot Wound Registry. The registry collects information from the University of Louisville Hospital, Kentucky Medical Examiner, County Coroners, Louisville Metro Police Department and crime lab. The registry included all victims of firearms-related injury, from 1/1/1996 and is current through 9/30/2005. Data from victims that were less than 21 years of age at their first visit for a firearms-related injury were recorded. Results: Thirty-two adolescents were identified as sustaining repeat firearmsrelated injuries. Thirty one (96.9%) were African American and all thirty-two victims were males. Eighty-four percent (27) of first time injuries were assaults, and sixty-nine percent of repeat incidents were assault (24 of 37). The average age of a victim at the time of first injury was 18.3 years, and 21.34 years for repeat incidents. Sixty-two percent (20) were uninsured at their first presentation. This increased to 80% for repeat incidents. Eleven of the 32 patients (34%) were students. Nine of the 32 adolescents (28.1%) died during a repeat firearm incident. Seven were due to homicide, one due to suicide, and one accidental death during a game of Russian Roulette. Toxicology results from all incidents showed 26 positive tests, 8 negative tests, and 35 unknown or not tested toxicology samples. THC was identified in 22 (84.6%) of the 26 positive tests, and cocaine was identified in 11 (42.3%) of the 26 positive tests. Opiates and Benzodiazepines were each positive in 8 (30.8%) of the 26 positive tests. Twenty-two of the 69 incidents (31.9%) occurred on roadways or sidewalks as “drive-by” shootings. Eight of the 69 incidents (11.6%) involved more than one patient in the shooting, but the majority involved only the adolescent (81.2%). Conclusion: 738 adolescents sustained firearms-related injuries at our urban trauma center during the time period of 1/01/96-9/30/2005. The number of firearms-related injuries in our adolescent patient population has been increasing for the past four years. The socioeconomic impact firearms have on our adolescent population is significant and is highlighted by our study of these thirty-two patients. Specifically, our study shows that the African American adolescent male and young adult population is especially at risk to gun violence in our urban setting. Preventive measures and out-reach programs must be directed at this high risk population if this trend is to be reversed.
70
Intimate Partner Violence Related Suicidality Is Lower in Hispanic Emergency Department Patients Than Non Hispanics
Larkin GL, Anglin D/UT Southwestern Medical School, Dallas, TX; USC - Keck School of Medicine, Los Angeles, CA
Study Objectives: To measure the association between intimate partner violence (IPV) and comorbid suicidality and underlying DSM IV Axis I disorders in a multiethnic, non-psychiatric emergency department population. Methods: Prospective cohort of ambulatory adult (⬎ 18 years) patients recruited during random time blocks using a bilingual, computerized, confidential, mental health screen in an urban, 145,000-visit ED. Patients presenting with overt mental health complaints were excluded. Uni- and multivariate comparisons were made with chi-square and logistic regression to assess differences of proportion between IPVexposed and IPV-naive patients. Results: The sample of 1,590 patients was representative of the overall ED population, with 29% white, 39% African American, and 32% Hispanic ethnicity; 845 (53%) were male. Three-month IPV was present in 219 (14.6%) but was significantly lower for Hispanics than non-Hispanics (11% v. 16.3%; p ⬍ .01). Acute IPV was less common in males (5.6% v. 9.4%) and Hispanics (6.1% v. 8.1%, p ⬍
S22 Annals of Emergency Medicine
.01). Of current IPV patients, 23% reported active suicidal ideation compared to 5.8% of non-IPV patients [OR: 4.8 (3.3, 7.1)]. Specific plans for suicide were present in 5.5% of IPV patients compared to only 1.1% of the overall sample [OR: 5.2 (2.4, 11.5)]. The rates of PTSD and Acute Stress Disorder (ASD) in IPV victims were 43.4% and 6.4%, significantly higher than the rates of PTSD (15.2%) and ASD (2.0%) in IPV-naive patients (p ⫽ .000). Hispanic IPV patients developed PTSD less commonly than non-Hispanics (13.9% v 37.6%; p ⫽ .000). Regardless of ethnicity, IPV victims reported significantly more at-risk drinking [19.6% v. 11.7%, OR: 1.8 (1.3, 2.7)] and Axis I DSM IV mood and anxiety disorders [78% v. 44%, OR: 4.6(3.3, 6.5)]. Conclusion: IPV is associated with significant levels of suicidal ideation and mental illness. However, the prevalences of IPV and IPV-associated psychiatric comorbidity are both lower in Hispanic than in non-Hispanic ED patients.
71
Effects of Holidays, Weekends, and Seasons on Intimate Partner Violence Incidents
Pawlowski JM, Moscati R, Bailey G, Priya A/University at Buffalo, Buffalo, NY; University at Buffalo, Buffalo, NY
Study Objectives: To observe trends in intimate partner violence (IPV) reports with special interest to the effects of holidays, weekends and seasons. Methods: Two independent databases consisting of the number of IPV reports by date were obtained. One database consisted of reports from the city police department from 01/98-12/02. The second database consisted of reports from the ED of the regional Level I trauma center from 7/97-12/02. Within each database compaisons were made between weekdays (Mon-Thu) and weekends (Fri-Sun), holidays and non-holidays, and summer (Apr-Sep) and winter (Oct-Mar). Odds ratios (OR) and 95% confidence intervals (CI) were calculated for the comparisons. Significant deviations in the number of IPV repors occurred where the comparisons yielded statistically different OR for the same comparison in both databases. Results: The police data ranged fom 0-34 reports/day, while the hospital data had a range of 0-5 reports/day. A significant increase in reports was seen on weekends when compared with weekdays in both sets of data. Although overall reporting was higher on holidays than non-holidays, the only holiday with a significant increase was New Years Day. The seasonal data revealed an increase in reports in the summer, this increase was significant only in the police data. Conclusion: Data taken from the two separate databases over the same 5 year time period revealed a significant increase in IPV only on weekends and New Years Day. Other comparisons showed less significant increases in only one database or neither. Common assumptions about the factors that may increase IPV such as time spent indoors and family holidays, did not hold true in our study.
72
Does Sex Matter? Effect of Gender in Intimate Partner Violence Screening
Gerlach LB, Datner EM, Hollander JE, Zogby KE, Robey JL, Wiebe DJ/ Department of Emergency Medicine University of Pennsylvania, Philadelphia, PA; Department of Biostatistics and Epidemiology University of Pennsylvania School of Medicine, Philadelphia, PA
Study Objectives: Over 250,000 patients are treated annually in United States Emergency Departments (EDs) for intimate partner violence (IPV) related injuries. While many studies have assessed the validity of existing screening measures, few studies have investigated the impact of gender of the screener on detection of IPV. The goal of this study is to determine if the gender of the screener influences patient self report of IPV. Methods: Study Design: We performed a prospective cross-sectional survey at an urban academic medical center with an ED census of 52,000 patients located in Philadelphia, PA. Participants: A consecutive sample of adults who presented to the ED from 07: 00-24:00, 7 days/week, over an 8-week period between January and March 2006
Volume , . : October
Research Forum Abstracts were screened for IPV. After verbal consent, screening was performed by 26 trained research assistants (11 women, 15 men). Patients were excluded if they were unable to answer questions, reported abuse from someone other than an intimate partner, or were unable to provide informed consent. Data Collection: IPV was detected using a validated 4-item tool adapted from the Abuse Assessment Scale to measure both physical and psychological abuse. An indication of being hit, kicked, punched, or threatened by an intimate partner within the past 6 months was considered a positive IPV screen. Data Analysis: Comparison of baseline demographic medical history and disposition was performed using chi-square tests and t-tests. Determination of whether the screening outcome varied by screener sex was assessed using chi-square tests. Logistic regression was used to control for patient age and adjust for nonindependence among screening staff. Results: A total of 3,349 patients participated (62.1% female; mean age 42 years; 67.5% African American, 22.9% white). Male and female patients did not differ significantly with respect to demographic information, chief complaint, or disposition. IPV in the past six months was reported for 52 female patients (2.5%) and 22 male patients (1.7%). The sex of the screener was not significantly associated with the screening outcomes for male (odds ratio ⫽ 1.20, CI ⫽ 0.52,2.79) or female patients (odds ratio ⫽ 0.91, CI ⫽ 0.51,1.61) (Table). Conclusion: Sex of the screener does not appear to affect disclosure of IPV. These results lend support for the continuation of existing screening practices.
73
Characteristics of Partner Violence Perpetrators in the Emergency Department
Johnson BD, Gasgarth R, Kemball RS, Houry D/Emory University, Atlanta, GA
While quantitative and qualitative data exists for victims of intimate partner violence (IPV) presenting to emergency departments, little is known about IPV perpetrators. Study Objectives: To describe demographic characteristics of perpetrators as well as to determine perpetrator perception of their IPV, perpetrator interest and involvement in re-education programs and IPV resources, and the perpetrator’s prior disclosure of IPV to other people. Methods: The study was conducted at the ED of a level 1 urban trauma center with an annual census of 105,000 patients. All adult patients aged 18-55, regardless of chief complaint, who presented to the ED during enrollment times of 8am to 5pm Monday thru Thursday during a 2 month period were eligible to participate. Patients were excluded if they were intoxicated, psychotic, didn’t speak English, or if they were medically unstable. Subjects were screened for IPV perpetration via a touch screen computer program that included questions on general health risk behaviors. Subjects that screened positive for perpetration were interviewed by a research assistant using a questionnaire. This questionnaire was developed and pilot tested with the assistance of Men Stopping Violence prior to use in our study. Three themes were included in the questionnaire: 1) perpetrator perception of their IPV, 2) interest and involvement in both re-education programs and IPV resources, and 3) the perpetrator’s communication of IPV with others. Results: 40 patients disclosed perpetration on the computer kiosk and participated in the survey. Age range was 18-51; mean and mode age was 33 years. Sixteen participants disclosed only perpetration behaviors and 24 disclosed both victimization and perpetration behaviors. The majority of perpetrators were African American (83%), male (58%), and disclosed illicit drug use (63%). 40% of perpetrators answered “yes” when asked if they think they abused a partner within the past 12 months; yet every perpetrator answered “no” to the question if they think abuse is okay. Only 5% of males and no females had ever enrolled in a re-education class for IPV and 24% of female and 9% of males had or were currently using IPV resources, yet 38% of perpetrators were interested in re-education programs. Female perpetrators preferred group class (65%) and male perpetrators preferred 1 on 1 counseling (65%). Only 18% had ever had any friends, family, or physicians ask about their perpetration (6% females, 26% males), whereas 40% reported they selfdisclosed perpetration (35% females, 43% males). Of those who disclosed
Volume , . : October
perpetration to someone, 75% reported the disclosure had influenced, made them think about, or changed their behavior. Conclusion: Perpetrators are willing to disclose IPV and many are interested in re-education classes or IPV resources. Emergency physicians should consider asking patients about both victimization and perpetration behaviors and referring patients as needed.
74
Substance Use and Violence Among Emergency Department Patients with Minor Injury
Prekker ME, Olsen JD, Hansen J, Miner JR/Hennepin County Medical Center, Minneapolis, MN
Study Objectives: Alcohol and illicit drug use increases the risk of violence-related injury. We determine the prevalence of substance use among ED patients with minor injury and examine the relationship between substance use and assault. Methods: Cross-sectional study at an urban, Level 1 Trauma Center. All patients who presented to the ED during randomly distributed shifts totaling 336 hours between June and August, 2005 were screened for trauma (N ⫽ 4246). Non-critically ill, adult patients with minor injury were enrolled in a cohort study. Patients with altered mental status (including acute intoxication) precluding informed consent were excluded. Alcohol and/or illicit drug users (N ⫽ 282) were identified from this larger study and the majority (86%, N ⫽ 243) completed an additional interview. Heavy alcohol users were defined as those with “frequent” use or ⬎ 5 drinks on one occasion. Data were analyzed using descriptive statistics and the chi-square test. Results: Of 434 patients with minor injury, 106 (24%) were heavy alcohol users and 74 (17%) reported illicit drug use (most commonly marijuana or cocaine). Thirty-seven of 106 (35%) heavy alcohol users also used drugs. Heavy alcohol users reported assault as contributing to their injury more frequently than patients with moderate or no alcohol use (26% vs. 15%, p ⫽ 0.006). A higher rate of assault was also observed among illicit drug users versus non-users (24% vs. 16%, p ⫽ 0.07). Of the 319 study patients with acute injury (ED visit within 24 hours), 34% reported drug or alcohol use over the preceding 24 hours. These recent users, as compared to non-users, were more frequently assaulted (27% vs. 13%, p ⫽ 0.003), uninsured (36% vs. 17%, p ⬍ 0.001) and lacking a primary care provider (63% vs. 50%, p ⫽ 0.01). Conclusion: Substance use is a significant contributor to violence-related injury. ED-based intervention may decrease this risk in a population without clear access to follow-up care.
75
Evaluation of In-Custody Deaths Proximal to Use of Conductive Energy Devices
Vilke G, Johnson W, Castillo EM, Ederheimer JA, Wexler C, Sloane CM, Chan TC/UCSD Medical Center, San Diego, CA; PERF, Washington DC, DC; SDSU, San Diego, CA
Study Objectives: Conductive Energy Devices (CEDs) are being used by many law enforcement agencies as a less lethal means of gaining control of subjects, particularly those who are armed, physically resistive or under the influence of illicit stimulant drugs that can offer “superhuman strength” where attempts to control are likely to result in injuries to officers or the subject himself. The Taser is the most commonly used CED, but others exist on the market. We sought to describe the clinically relevant medical circumstances surrounding the deaths in subjects who died while in police custody after being exposed to a CED activation. Methods: This study is an extensive survey of all police agencies that had a reported CED related death. The list of deaths was obtained by cross referencing the published lists from Amnesty International and the Arizona Republic occurring between September, 1999 and May 31, 2005. A survey tool was developed by a group of independent panelists with experience in CED use which included senior law enforcement officials in order to ensure that the language used that would solicit the same information across a national setting. Law enforcement agencies were asked to participate in the data collection via a secure website or paper format that could be faxed back. Medical examiner reports were obtained when available. The participation was voluntary and the information received via the web base was blinded. Descriptive statistics were utilized to clinically relevant medical circumstances surrounding the deaths in subjects. Results: There were 118 unique proximity deaths in subjects who had CED activations identified. These cases involved a total of 96 law enforcement jurisdictions in 26 different states. Each of these agencies was surveyed, with 60 (63%) agencies responding, resulting in data from 65% (77 of 118) of the CED proximity deaths.
Annals of Emergency Medicine S23
69
Recurring Adolescent Victims of Firearms-Related Injuries: A Population at Risk
Cox JA, Davidson DJ, Smock WS/University of Louisville, School of Medicine, Louisville, KY
Introduction: Adolescents that sustain a firearms-related injury are at increased risk of being repeat victims of gun-violence. Nearly 5% of adolescent firearms-related victims presenting to an urban trauma center will sustain additional firearms-related injuries. Study Objective: This study examined the characteristics of repeat gunshot wound victims that were adolescents at time of first injury. Methods: Data was reviewed from the University of Louisville Gunshot Wound Registry. The registry collects information from the University of Louisville Hospital, Kentucky Medical Examiner, County Coroners, Louisville Metro Police Department and crime lab. The registry included all victims of firearms-related injury, from 1/1/1996 and is current through 9/30/2005. Data from victims that were less than 21 years of age at their first visit for a firearms-related injury were recorded. Results: Thirty-two adolescents were identified as sustaining repeat firearmsrelated injuries. Thirty one (96.9%) were African American and all thirty-two victims were males. Eighty-four percent (27) of first time injuries were assaults, and sixty-nine percent of repeat incidents were assault (24 of 37). The average age of a victim at the time of first injury was 18.3 years, and 21.34 years for repeat incidents. Sixty-two percent (20) were uninsured at their first presentation. This increased to 80% for repeat incidents. Eleven of the 32 patients (34%) were students. Nine of the 32 adolescents (28.1%) died during a repeat firearm incident. Seven were due to homicide, one due to suicide, and one accidental death during a game of Russian Roulette. Toxicology results from all incidents showed 26 positive tests, 8 negative tests, and 35 unknown or not tested toxicology samples. THC was identified in 22 (84.6%) of the 26 positive tests, and cocaine was identified in 11 (42.3%) of the 26 positive tests. Opiates and Benzodiazepines were each positive in 8 (30.8%) of the 26 positive tests. Twenty-two of the 69 incidents (31.9%) occurred on roadways or sidewalks as “drive-by” shootings. Eight of the 69 incidents (11.6%) involved more than one patient in the shooting, but the majority involved only the adolescent (81.2%). Conclusion: 738 adolescents sustained firearms-related injuries at our urban trauma center during the time period of 1/01/96-9/30/2005. The number of firearms-related injuries in our adolescent patient population has been increasing for the past four years. The socioeconomic impact firearms have on our adolescent population is significant and is highlighted by our study of these thirty-two patients. Specifically, our study shows that the African American adolescent male and young adult population is especially at risk to gun violence in our urban setting. Preventive measures and out-reach programs must be directed at this high risk population if this trend is to be reversed.
70
Intimate Partner Violence Related Suicidality Is Lower in Hispanic Emergency Department Patients Than Non Hispanics
Larkin GL, Anglin D/UT Southwestern Medical School, Dallas, TX; USC - Keck School of Medicine, Los Angeles, CA
Study Objectives: To measure the association between intimate partner violence (IPV) and comorbid suicidality and underlying DSM IV Axis I disorders in a multiethnic, non-psychiatric emergency department population. Methods: Prospective cohort of ambulatory adult (⬎ 18 years) patients recruited during random time blocks using a bilingual, computerized, confidential, mental health screen in an urban, 145,000-visit ED. Patients presenting with overt mental health complaints were excluded. Uni- and multivariate comparisons were made with chi-square and logistic regression to assess differences of proportion between IPVexposed and IPV-naive patients. Results: The sample of 1,590 patients was representative of the overall ED population, with 29% white, 39% African American, and 32% Hispanic ethnicity; 845 (53%) were male. Three-month IPV was present in 219 (14.6%) but was significantly lower for Hispanics than non-Hispanics (11% v. 16.3%; p ⬍ .01). Acute IPV was less common in males (5.6% v. 9.4%) and Hispanics (6.1% v. 8.1%, p ⬍
S22 Annals of Emergency Medicine
.01). Of current IPV patients, 23% reported active suicidal ideation compared to 5.8% of non-IPV patients [OR: 4.8 (3.3, 7.1)]. Specific plans for suicide were present in 5.5% of IPV patients compared to only 1.1% of the overall sample [OR: 5.2 (2.4, 11.5)]. The rates of PTSD and Acute Stress Disorder (ASD) in IPV victims were 43.4% and 6.4%, significantly higher than the rates of PTSD (15.2%) and ASD (2.0%) in IPV-naive patients (p ⫽ .000). Hispanic IPV patients developed PTSD less commonly than non-Hispanics (13.9% v 37.6%; p ⫽ .000). Regardless of ethnicity, IPV victims reported significantly more at-risk drinking [19.6% v. 11.7%, OR: 1.8 (1.3, 2.7)] and Axis I DSM IV mood and anxiety disorders [78% v. 44%, OR: 4.6(3.3, 6.5)]. Conclusion: IPV is associated with significant levels of suicidal ideation and mental illness. However, the prevalences of IPV and IPV-associated psychiatric comorbidity are both lower in Hispanic than in non-Hispanic ED patients.
71
Effects of Holidays, Weekends, and Seasons on Intimate Partner Violence Incidents
Pawlowski JM, Moscati R, Bailey G, Priya A/University at Buffalo, Buffalo, NY; University at Buffalo, Buffalo, NY
Study Objectives: To observe trends in intimate partner violence (IPV) reports with special interest to the effects of holidays, weekends and seasons. Methods: Two independent databases consisting of the number of IPV reports by date were obtained. One database consisted of reports from the city police department from 01/98-12/02. The second database consisted of reports from the ED of the regional Level I trauma center from 7/97-12/02. Within each database compaisons were made between weekdays (Mon-Thu) and weekends (Fri-Sun), holidays and non-holidays, and summer (Apr-Sep) and winter (Oct-Mar). Odds ratios (OR) and 95% confidence intervals (CI) were calculated for the comparisons. Significant deviations in the number of IPV repors occurred where the comparisons yielded statistically different OR for the same comparison in both databases. Results: The police data ranged fom 0-34 reports/day, while the hospital data had a range of 0-5 reports/day. A significant increase in reports was seen on weekends when compared with weekdays in both sets of data. Although overall reporting was higher on holidays than non-holidays, the only holiday with a significant increase was New Years Day. The seasonal data revealed an increase in reports in the summer, this increase was significant only in the police data. Conclusion: Data taken from the two separate databases over the same 5 year time period revealed a significant increase in IPV only on weekends and New Years Day. Other comparisons showed less significant increases in only one database or neither. Common assumptions about the factors that may increase IPV such as time spent indoors and family holidays, did not hold true in our study.
72
Does Sex Matter? Effect of Gender in Intimate Partner Violence Screening
Gerlach LB, Datner EM, Hollander JE, Zogby KE, Robey JL, Wiebe DJ/ Department of Emergency Medicine University of Pennsylvania, Philadelphia, PA; Department of Biostatistics and Epidemiology University of Pennsylvania School of Medicine, Philadelphia, PA
Study Objectives: Over 250,000 patients are treated annually in United States Emergency Departments (EDs) for intimate partner violence (IPV) related injuries. While many studies have assessed the validity of existing screening measures, few studies have investigated the impact of gender of the screener on detection of IPV. The goal of this study is to determine if the gender of the screener influences patient self report of IPV. Methods: Study Design: We performed a prospective cross-sectional survey at an urban academic medical center with an ED census of 52,000 patients located in Philadelphia, PA. Participants: A consecutive sample of adults who presented to the ED from 07: 00-24:00, 7 days/week, over an 8-week period between January and March 2006
Volume , . : October
Research Forum Abstracts were screened for IPV. After verbal consent, screening was performed by 26 trained research assistants (11 women, 15 men). Patients were excluded if they were unable to answer questions, reported abuse from someone other than an intimate partner, or were unable to provide informed consent. Data Collection: IPV was detected using a validated 4-item tool adapted from the Abuse Assessment Scale to measure both physical and psychological abuse. An indication of being hit, kicked, punched, or threatened by an intimate partner within the past 6 months was considered a positive IPV screen. Data Analysis: Comparison of baseline demographic medical history and disposition was performed using chi-square tests and t-tests. Determination of whether the screening outcome varied by screener sex was assessed using chi-square tests. Logistic regression was used to control for patient age and adjust for nonindependence among screening staff. Results: A total of 3,349 patients participated (62.1% female; mean age 42 years; 67.5% African American, 22.9% white). Male and female patients did not differ significantly with respect to demographic information, chief complaint, or disposition. IPV in the past six months was reported for 52 female patients (2.5%) and 22 male patients (1.7%). The sex of the screener was not significantly associated with the screening outcomes for male (odds ratio ⫽ 1.20, CI ⫽ 0.52,2.79) or female patients (odds ratio ⫽ 0.91, CI ⫽ 0.51,1.61) (Table). Conclusion: Sex of the screener does not appear to affect disclosure of IPV. These results lend support for the continuation of existing screening practices.
73
Characteristics of Partner Violence Perpetrators in the Emergency Department
Johnson BD, Gasgarth R, Kemball RS, Houry D/Emory University, Atlanta, GA
While quantitative and qualitative data exists for victims of intimate partner violence (IPV) presenting to emergency departments, little is known about IPV perpetrators. Study Objectives: To describe demographic characteristics of perpetrators as well as to determine perpetrator perception of their IPV, perpetrator interest and involvement in re-education programs and IPV resources, and the perpetrator’s prior disclosure of IPV to other people. Methods: The study was conducted at the ED of a level 1 urban trauma center with an annual census of 105,000 patients. All adult patients aged 18-55, regardless of chief complaint, who presented to the ED during enrollment times of 8am to 5pm Monday thru Thursday during a 2 month period were eligible to participate. Patients were excluded if they were intoxicated, psychotic, didn’t speak English, or if they were medically unstable. Subjects were screened for IPV perpetration via a touch screen computer program that included questions on general health risk behaviors. Subjects that screened positive for perpetration were interviewed by a research assistant using a questionnaire. This questionnaire was developed and pilot tested with the assistance of Men Stopping Violence prior to use in our study. Three themes were included in the questionnaire: 1) perpetrator perception of their IPV, 2) interest and involvement in both re-education programs and IPV resources, and 3) the perpetrator’s communication of IPV with others. Results: 40 patients disclosed perpetration on the computer kiosk and participated in the survey. Age range was 18-51; mean and mode age was 33 years. Sixteen participants disclosed only perpetration behaviors and 24 disclosed both victimization and perpetration behaviors. The majority of perpetrators were African American (83%), male (58%), and disclosed illicit drug use (63%). 40% of perpetrators answered “yes” when asked if they think they abused a partner within the past 12 months; yet every perpetrator answered “no” to the question if they think abuse is okay. Only 5% of males and no females had ever enrolled in a re-education class for IPV and 24% of female and 9% of males had or were currently using IPV resources, yet 38% of perpetrators were interested in re-education programs. Female perpetrators preferred group class (65%) and male perpetrators preferred 1 on 1 counseling (65%). Only 18% had ever had any friends, family, or physicians ask about their perpetration (6% females, 26% males), whereas 40% reported they selfdisclosed perpetration (35% females, 43% males). Of those who disclosed
Volume , . : October
perpetration to someone, 75% reported the disclosure had influenced, made them think about, or changed their behavior. Conclusion: Perpetrators are willing to disclose IPV and many are interested in re-education classes or IPV resources. Emergency physicians should consider asking patients about both victimization and perpetration behaviors and referring patients as needed.
74
Substance Use and Violence Among Emergency Department Patients with Minor Injury
Prekker ME, Olsen JD, Hansen J, Miner JR/Hennepin County Medical Center, Minneapolis, MN
Study Objectives: Alcohol and illicit drug use increases the risk of violence-related injury. We determine the prevalence of substance use among ED patients with minor injury and examine the relationship between substance use and assault. Methods: Cross-sectional study at an urban, Level 1 Trauma Center. All patients who presented to the ED during randomly distributed shifts totaling 336 hours between June and August, 2005 were screened for trauma (N ⫽ 4246). Non-critically ill, adult patients with minor injury were enrolled in a cohort study. Patients with altered mental status (including acute intoxication) precluding informed consent were excluded. Alcohol and/or illicit drug users (N ⫽ 282) were identified from this larger study and the majority (86%, N ⫽ 243) completed an additional interview. Heavy alcohol users were defined as those with “frequent” use or ⬎ 5 drinks on one occasion. Data were analyzed using descriptive statistics and the chi-square test. Results: Of 434 patients with minor injury, 106 (24%) were heavy alcohol users and 74 (17%) reported illicit drug use (most commonly marijuana or cocaine). Thirty-seven of 106 (35%) heavy alcohol users also used drugs. Heavy alcohol users reported assault as contributing to their injury more frequently than patients with moderate or no alcohol use (26% vs. 15%, p ⫽ 0.006). A higher rate of assault was also observed among illicit drug users versus non-users (24% vs. 16%, p ⫽ 0.07). Of the 319 study patients with acute injury (ED visit within 24 hours), 34% reported drug or alcohol use over the preceding 24 hours. These recent users, as compared to non-users, were more frequently assaulted (27% vs. 13%, p ⫽ 0.003), uninsured (36% vs. 17%, p ⬍ 0.001) and lacking a primary care provider (63% vs. 50%, p ⫽ 0.01). Conclusion: Substance use is a significant contributor to violence-related injury. ED-based intervention may decrease this risk in a population without clear access to follow-up care.
75
Evaluation of In-Custody Deaths Proximal to Use of Conductive Energy Devices
Vilke G, Johnson W, Castillo EM, Ederheimer JA, Wexler C, Sloane CM, Chan TC/UCSD Medical Center, San Diego, CA; PERF, Washington DC, DC; SDSU, San Diego, CA
Study Objectives: Conductive Energy Devices (CEDs) are being used by many law enforcement agencies as a less lethal means of gaining control of subjects, particularly those who are armed, physically resistive or under the influence of illicit stimulant drugs that can offer “superhuman strength” where attempts to control are likely to result in injuries to officers or the subject himself. The Taser is the most commonly used CED, but others exist on the market. We sought to describe the clinically relevant medical circumstances surrounding the deaths in subjects who died while in police custody after being exposed to a CED activation. Methods: This study is an extensive survey of all police agencies that had a reported CED related death. The list of deaths was obtained by cross referencing the published lists from Amnesty International and the Arizona Republic occurring between September, 1999 and May 31, 2005. A survey tool was developed by a group of independent panelists with experience in CED use which included senior law enforcement officials in order to ensure that the language used that would solicit the same information across a national setting. Law enforcement agencies were asked to participate in the data collection via a secure website or paper format that could be faxed back. Medical examiner reports were obtained when available. The participation was voluntary and the information received via the web base was blinded. Descriptive statistics were utilized to clinically relevant medical circumstances surrounding the deaths in subjects. Results: There were 118 unique proximity deaths in subjects who had CED activations identified. These cases involved a total of 96 law enforcement jurisdictions in 26 different states. Each of these agencies was surveyed, with 60 (63%) agencies responding, resulting in data from 65% (77 of 118) of the CED proximity deaths.
Annals of Emergency Medicine S23