320: Association between Socioeconomic Status and the Incidence and Severity of Traumatic Brain Injury

Research Forum Abstracts personal and environmental correlates of injury risks to pedestrians including traffic flow, density, education level, English fluency and socioeconomic status (SES). Pedestrian injuries have a greater negative impact on those of lower SES. Previous studies show that disparities exist among populations dependent on SES with regard to the incidence and outcomes of pedestrian crashes. The purpose of this study is to explore the relationship between personal and environmental risk factors within a large economically and culturally diverse county in Southern California. Methods: Pedestrian crash data were analyzed from the California Statewide Integrated Traffic Records System (SWITRS) (2000-2004). A Geographic Information System was used to identify the location of pedestrian crashes and this was overlaid onto demographic and personal characteristic data from the US Census (2000). The relationship between population characteristics and pedestrian crash counts in each of the 577 census tracts was assessed using negative binomial regression controlling for population size. Results: 4,000 pedestrian crashes were identified and geo-coded in a county population of 2,846,289. The percent of the population living in households with low income (less than 185% of the federal poverty level) was the strongest predictor of pedestrian injuries. One fourth of census tracts had less than 8.7% of residents with low income and averaged 11 per 100,000 pedestrian crashes annually. One fourth of the census tracts had more than 32.2% of residents with low income, and an average of 44 pedestrian crashes per 100,000 annually. Negative binomial regression showed low income was the strongest predictor of pedestrian injuries: each 1% increase in the percent of resident with low income was associated with a 2.8% increase in pedestrian crashes. The percent of residents age 14 yrs or less, adult residents who had not completed high school, residents who spoke English less than ”very well” and spoke another language at home, and the population density were each associated with a higher frequency of pedestrian crashes. However, when low income was added to these four regression models, the relationship between low income and pedestrian crashes increased. Conclusion: Pedestrian crashes are four times more frequent in poor neighborhoods. Neither age of the population, education, English fluency, nor population density explained the effect of poverty. The results demonstrate significant geographic and demographic based disparities in pedestrian crashes. Further understanding of the complex social and ecological forces linked to pedestrian crashes can help direct educational programs, interventions, policy and ultimately improve injury disparities.

318

Injured and Uninjured ED Patients: Alcohol and Marijuana Use, Risky Behavior, and Readiness to Change

Becker BM, Woolard R, Baird J, Mello MJ, Nirenberg T, Lee C, Longabaugh R/ Rhode Island Hospital, Providence, RI; Brown University, Providence, RI

Study Objectives: In currently published studies, only patients with injury have been selected as targets for ED-based brief interventions (BI) addressing alcohol use, and other risky behaviors. These patients have a profile of risk (past and future potential for injuries; risky alcohol and other substance use) and are open to a BI. In this study, investigators enrolled injured and non-injured ED patients, evaluating their risk profile and readiness to change. Methods: We recruited 515 sub-critically injured and uninjured adults seeking treatment in an ED who reported alcohol and/or marijuana use in the last month, were English speaking, and lived within 2 hours of the hospital. Participants were consented and completed instruments examining their substance use, prior injuries, high risk behaviors, and readiness to change alcohol and/or marijuana use. Analyses were conducted using SAS; raw scores on assessments were transformed prior to analysis if appropriate. The hospital IRB approved all protocols. Results: Of 515 participants, 345 (67%) were men, with a mean age of 28 (range: 18-55yrs), and 345 (68%) were white. The mean AUDIT score was 11.6 (SD 8.1) and the mean number of days of THC use (in last 30 days) was 12.6 (SD⫽11.7). In the three months prior to enrollment, 119 reported driving after binging on alcohol (23%) and 149 (29%) reported being involved with a fight. There was no significant difference between the injured or uninjured participants for any of these behaviors. More than 80% of injured and uninjured participants reported having at least 1 injury in the past 12 months. More than 80% of them reported psychosocial problems associated with their alcohol use and 64% reported problems associated with their THC use. There was no difference in the frequency of previous injuries between injured and participants. Sixty-nine percent of injured and 71% of uninjured participants reported readiness to change their conjoint use of both alcohol and THC.

S100 Annals of Emergency Medicine

Conclusion: Sub-critically injured and uninjured ED patients in this study demonstrated similar profiles of risk behaviors, and were equally ready to change their substance use. Both injured and uninjured ED patients should be queried about substance use and risky behaviors, and should be offered behavioral intervention.

319

The Relationship Between Facial Bone Fractures With Accompanied Injuries and the Helmet Types of the Motorcyclists in Traffic Accidents

Liu CH, Huang YC/Chia-Yi Christian Hospital, Chia-Yi, Taiwan

Study Objectives: The motorcycle is popular and is also a necessary motor vehicle in Asian countries. Although the mortality of head trauma decreased significantly since 1987 because helmet use was mandated by the law, motorcyclists remain suffering from facial, orbital trauma, and related intracranial hemorrhages from motor vehicle accidents. Various types of helmets are marketed with different levels of protection. We sought the relationship between helmet types and the injuries of head and face resulted from motor vehicle accidents. Methods: A retrospective cohort study was undertaken with motorcyclists suffering facial bone fractures being retrieved from our trauma registration from January 1st to December 31st in 2006. Helmets were classified into 3 groups, Group1: full-face helmet, Group 2: open-face helmet with a shield and Group 3: open-face helmet without a shield. Location and severity of facial bone fractures, ocular injuries, and intracranial hemorrhages were recorded. Protection effects of each type of helmet were analyzed from the relative risks to multiple facial bone fractures, ocular injuries, and intracranial hemorrhages. Results: After exclusion of patients without complete records, 115 patients were enrolled. There were 77 men and 38 women with age of 38.9 ⫾ 18.2 years-old. Group 1 had only 6 patients (5.2%), group 2 had 41 patients (35.7%) and group 3 had 68 patients (59.1%). The injury rate of facial bones were maxilla-zygoma (52.8%), mandible (16.8%), nasal (16.8%), orbital bone (13.6%), others including frontal bone (21.6%) and multiple facial bones involved (19.2%). The fracture sites in group 1 include 3 (50%) nasal bone, 2 maxilla-zygoma, 1 orbital bone and the injuries are relatively simple and mild. The presence of multiple fracture sites were none in group 1, 8 (19.5%) in group 2 and 13 (19.1%) in group 3. The associated intracranial hemorrhage are also none in group 1, 7 (17.1%) in group 2 and 15 (22.1%) in group 3. There accompanied ocular injuries are 1 (11.1%) in group 1, 11 (26.8%) in group 2 and 13 (19.1%) in group 3. Conclusion: The majority of the facial fractures are caused by less or none facial protective helmets. Open-face helmets provide less protection to the face during motorcycle accident. Risks of facial bone fractures, ocular injuries, and even intracranial hemorrhages were higher in patients whose helmet provided less protection. There were rare cases of facial bone fractures and associated injuries in motorcyclists with full-face helmets in motorcycle crashes. Although wearing the fullface helmet is hot, humid, and inconvenient, they are all worthy from the greater protection to face, eyes, and brain.

320

Association between Socioeconomic Status and the Incidence and Severity of Traumatic Brain Injury

Shin S, Ahn K, Kwak Y, Song K, Park J, Youn Y/Seoul National University College of Medicine, Seoul, Republic of Korea

Study Objectives: The association between the socioeconomic status (SES) and the incidence and severity of traumatic brain injury (TBI) were evaluated through a population-based prospective study in Korea. Methods: All members (approximately 47 million) of the compulsory Korean National Health System (KNHS) were initially included. The SES was classified according to the individual premium of the KNHS. The KNHS premium is determined either by the salary level (Employee Insurance, EI) or property owned (Self-employed Insurance, SEI). The medical aid program (MA), where the premium is exempted, was also included. The data for the patients with a TBI were obtained from the National Injury Database (NIDB) from 2001 to 2003. The age-sex standardized incidence rates (SIRs) and severity, which were derived from the Excess Mortality Ratio-adjusted International Classification of Disease 10th Edition-based Injury Severity Score (EMR-ISS) and is similar to the New Injury Severity Score (NISS), were estimated. The severity was categorized as mild, moderate, severe, and critical according to the EMR-ISS. The association

Volume , .  : September 

Research Forum Abstracts between the SES and critical TBI was evaluated using a multivariate logistic regression model and Cox-proportional hazard regression analysis. Results: The total number of TBI was 4,148,640. SIRs and relative risks decreased with increasing SES from the MA population to the highest quintile in both EI and SEI groups. The adjusted odds ratios for the incidence of a more severe TBI (EMR-ISS⬎⫽25) decreased with increasing SES. Adjusted hazard ratios (HRs) revealed a negative dose-response relationship between mortality and SES. The adjusted HR of the highest SES populations compared with the MA population for a critical TBI (death) were 0.47 (95% confidence interval, 0.44⬃0.51) in the SEI population and 0.51 (95% confidence interval, 0.48⬃0.55) in the EI population, respectively. Conclusion: The SIRs of a TBI were significantly lower in the higher SES populations than in the lowest SES group. The incidence of a critical TBI was highest in those with the lowest SES.

321

Motorcycle Helmet Use in the Dominican Republic Kessler D/Mount Sinai Medical Center, New York, NY

Study Objectives: Road traffic injuries are a growing global public health problem, resulting in over 1.2 million deaths worldwide and 50 million injuries every year. Low- and middle-income countries (LMICs) are disproportionately affected, bearing an economic burden as high as 2% of the gross national product according to some studies. Motorcycles are increasingly more common in many LMICs as a cheap and available means of transportation. Motorcycle accidents often result in serious morbidity and mortality. It is well established that motorcycle helmet use significantly reduces both mortality as well as the economic burden imposed by motorcycle crashes. In the Dominican Republic, there is few data surrounding motorcycle helmet use, yet motorcycle crashes have been recognized as the greatest cause of mortality due to road traffic injuries. This objective of this study is to investigate the rate of helmet use among both drivers and passengers of motorcycles int he Dominican Republic. Methods: This is an observational study of helmet use. Observations were made of motorcycle riders during four separate 30 minute intervals at two different locations in the city of Santiago for a total of 2 hours. Helmets was defined as any hard protective covering. Variables were recorded for any rider entering the intersection: gender, helmet use, rider position (driver or passenger). Results: 1152 motorcycle riders were observed, 980 drivers and 172 passengers. The majority of riders were male (976) and 49 were female. Overall 16% of riders observed were wearing helmets (183), and 100% of those were drivers. No females or passengers were observed to be wearing motorcycle helmets. Close to 10 riders per minute were observed during this study, 8 of whom did not wear helmets. Conclusion: Motorcycle helmet use is very low in the Dominican Republic., particularly among passengers and females. A public health campaign to increase motorcycle helmet compliance could have significant impact on morbidity, mortality and the economic burden associated with such injuries.

322

Project Integrate: Translating Screening and Brief Interventions for Alcohol to a Community Emergency Department

Mello MJ, Smith J, Baird J, Nirenberg T, Dinwoodie R/Rhode Island Hospital, Brown University, Providence, RI; Rhode Island Hospital, Providence, RI; Kent Hospital, Warwick, RI

Study Objectives: Screening and brief intervention (SBI) for alcohol in the emergency department has been proven to be effective by numerous research studies but has not been universally implemented. The objective of this study was to examine the translation of SBI into a busy community ED environment that does not currently use a standardized alcohol screening tool or have a protocol for targeted brief interventions. Our aims included determining if integrating a protocol for screening for high risk alcohol use: 1) increases screening by providers; 2) increases referral to an interventionist in the ED for BI 3) increases screening and referral for alcohol treatment when an interventionist is not present in the ED. Methods: Intervention ED was a community hospital with 60,000 ED visits/year where physicians worked in a fee for service reimbursement practice. A research assistant (RA) trained in MI was present in a discrete area of the ED for non-life threatening problems on selected day/evening shifts encompassing all days of the week from November 2005 to May 2006. ED staff were given a brief training on ACEP/NHTSA screening tool (NIAAA quantity/frequency guidelines ⫹ CAGE).

Volume , .  : September 

Patients were given screening tool to self administer on arrival in ED, and data was collected on frequency of physician use of screening tool, and accuracy of referral with qualified patients to RA for a brief targeted alcohol intervention. To examine for change in screening practices when the RA was not in the ED, a retrospective explicit review of a minimum of 254 ED charts noting frequency of screening and any referrals was also done 1-month prior to and 1-month post ED intervention. IRB of both the community hospital and research institution approved all protocols. Project Integrate was funded by a grant from Emergency Medicine Foundation. Results: Screening data results: While the RA was on duty, 90% (834) of eligible patients in the discrete ED area completed screening tool, and 89% of screens were evaluated by ED staff. Over the study period, accuracy of ED staff identification of positive screens and referral to the RA was 88% overall. There was a trend toward improvement of accuracy of physicians in identifying positive screens over the study period but it was not statistically significant (F(6,187 ⫽ 1.83; p⫽ .09). By individual physician there was also a trend toward improvement in utilization of screening tool from physician’s first intervention shift to the third, but this was not statistically significant (F(2.69)⫽0.57: p⫽ ⬎.10). Chart Review Results: Documented screening for alcohol use in ED chart was 50% 1 month before and 1 month after the intervention. Identified barriers included: physician’s reluctance to attend training; initial nurse leadership refusal to participate, pressure to improve patient flow. Identified opportunities included: nursing interest after observation of SBI; 15 nurses trained in SBI after study intervention concluded. Conclusion: With minimal training, support for distribution of screening tools, and trained staff present for referrals, high levels of screening for alcohol problems and appropriate referral can occur in a community ED. However, important barriers to translating from research to standard clinical practice exist and additional resources, such as the presence of a dedicated interventionist may be needed.

323

Effect of an Educational Intervention on the Knowledge and Use of Child Car Safety Restraints in Parents of 4-12-Year-Old Children

Jackson AT, Andrada ER, Woods R, Morris C, Panacek EA/University of California-Davis, Sacramento, CA

Study Objectives: The American Academy of Pediatrics (AAP) has made recommendations for optimal child car safety restraint use that have been shown to reduce the risk of injury to children in motor vehicle accidents. We sought to determine the knowledge of and compliance with the AAP child car safety recommendations among parents of 4-12 year-old children before and after a brief educational intervention. Methods: We evaluated a convenience sample of parents with 4-12-year-old children at regional children’s events. Eligible parents completed a written preintervention survey assessing their baseline knowledge of the American Academy of Pediatrics child car safety recommendations as well as their baseline child car safety restraint practices. The parents then participated in a brief educational intervention involving watching a video presentation and reading handouts discussing the AAP recommendations before completing a written survey assessing their immediate post-intervention knowledge of the recommendations. Finally, the parents completed a phone follow-up survey between 2 and 6 months after the intervention that assessed their delayed post-intervention knowledge of the AAP recommendations and their delayed post-intervention child car safety restraint practices. Simple descriptive summary statistical analysis with 95% CIs was performed. Results: A total of 101 parents and 151 children were enrolled at 3 different events. There were statistically significant improvements in the percentages of correct answers to all 3 of the AAP knowledge questions from the pre-intervention survey (36%, 23% and 54%) to the immediate post-intervention survey (84%, 97% and 79%). However, the percentages of correct answers to the delayed post-intervention survey (60%, 23% and 66%) were not significantly different from the preintervention survey. The forward-facing car seat compliance rate had a statistically significant improvement from 87% to 98%. The booster seat compliance rate improved from 50% to 65%, but this was not statistically significant. Conclusion: Our educational intervention was associated with improved knowledge of the AAP recommendations initially, but not 2-6 months later. This suggests that parents can learn from educational interventions, but that reinforcement of the information may be necessary in order for long-term retention to occur. Our

Annals of Emergency Medicine S101