Completeness of cause of injury coding in healthcare administrative databases in the United States, 2001.

OBJECTIVES: To determine the completeness of external cause of injury coding (E-coding) within healthcare administrative databases in the United States and to identify factors that contribute to variations in E-code reporting across states. DESIGN: Cross sectional analysis of the 2001 Healthcare Cost and Utilization Project (HCUP), including 33 State Inpatient Databases (SID), a Nationwide Inpatient Sample (NIS), and nine State Emergency Department Databases (SEDD). To assess state reporting practices, structured telephone interviews were conducted with the data organizations that participate in HCUP. RESULTS: The percent of injury records with an injury E-code was 86% in HCUP's nationally representative database, the NIS. For the 33 states represented in the SID, completeness averaged 87%, with more than half of the states reporting E-codes on at least 90% of injuries. In the nine states also represented in the SEDD, completeness averaged 93%. Twenty two states had mandates for E-code reporting, but only eight had provisions for enforcing the mandates. These eight states had the highest rates of E-code completeness. CONCLUSIONS: E-code reporting in administrative databases is relatively complete, but there is significant variation in completeness across the states. States with mandates for the collection of E-codes and with a mechanism to enforce those mandates had the highest rates of E-code reporting. Nine statewide ED data systems demonstrate consistently high E-coding completeness.

An evaluation of a system-change training model to improve emergency department response to battered women.

OBJECTIVES: To evaluate a system-change model of training from the Family Violence Prevention Fund and the Pennsylvania Coalition Against Domestic Violence for improving the effectiveness of emergency department (ED) response to intimate partner violence (IPV). METHODS: An experimental design with outcomes measured at baseline, 9-12, and 18-24 months post-intervention. Twelve hospitals in Pennsylvania and California with 20,000-40,000 annual ED visits were randomly selected and randomly assigned to experimental and control conditions. Emergency department teams (physician, nurse, social worker) from each experimental hospital and a local domestic violence advocate participated in a two-day didactic information and team planning intervention. RESULTS: The experimental hospitals were significantly higher than the control hospitals on a staff knowledge and attitude measure (F = 5.57, p = 0.019), on all components of the "culture of the ED" system-change indicator (F = 5.72, p = 0.04), and in patient satisfaction (F = 15.43, p < 0.001) after the intervention. There was no significant difference in the identification rates of battered women (F = 0.411, p = 0.52) (although the linear comparison was in the expected direction) in the medical records of the experimental and control hospitals. CONCLUSIONS: A system-change model of IPV ED training was effective in improving staff attitudes and knowledge about battered women and in protocols and staff training, as well as patient information and satisfaction. However, change in actual clinical practice was more difficult to achieve and may be influenced by institutional policy.

Errors in emergency medicine: research strategies.

Medical errors in emergency departments (EDs) may be an important "public health risk." Therefore, scientific public health approaches should be used to 1) assess the magnitude of emergency medical errors with surveillance methods, 2) identify causal factors of these medical errors with clinical epidemiologic methods, and 3) evaluate the effectiveness of interventions aimed at reducing or eliminating emergency medicine errors with health service research techniques. Since errors result from complex human-system interaction, research efforts should focus on actions taken by the patient, factors concerning the ED environment, and actions taken by health care workers. Other medical and nonmedical fields have already made great advancements in studying and reducing human error. Many of these advancements could readily be adapted to study emergency medical errors.

Traumatic brain injuries evaluated in U.S. emergency departments, 1992-1994.

OBJECTIVE: To describe the incidence and patient characteristics of traumatic brain injuries (TBIs) treated in U.S. emergency departments (EDs). METHODS: A secondary analysis was performed on data from the National Hospital Ambulatory Medical Care Survey administered from 1992 to 1994. An ED visit was determined to represent a case of TBI if the case record contained ICD-9-CM codes of 800.0-801.9, 803.0-804.9, or 850.0-854.1. RESULTS: The average annual estimate of new TBI treated in U.S. EDs was 1,144,807, equaling 444 per 100,000 persons (95% CI = 390 to 498), which represents approximately 3,136 new cases of TBI per day and accounts for 1.3% of all ED visits. Males were 1.6 times as likely as females to suffer TBI until the age of 65 years, when the female rate exceeded the male. The rate for blacks was 35% higher than that for whites. The highest overall incidence rate of TBI occurred in the less-than-5-year age group (1,091 per 100,000), closely followed by the more-than-85-year age group (1,026 per 100,000). Falls represented the most common mechanism of TBI injury, followed by motor vehicle-related trauma. CONCLUSIONS: This study underscores the ongoing need for effective surveillance of all types of TBI and evaluation of prevention strategies targeting high-risk individuals. It serves as a clinically grounded and ED-based corroboration of prior survey research, providing a basis for comparison of incidence rates over time and a tool with which to measure the efficacy of future interventions.

Effectiveness of ignition interlock devices in reducing drunk driving recidivism.

OBJECTIVE: To determine if ignition interlock devices reduce driving while intoxicated (DWI) recidivism. SEARCH STRATEGIES: Cochrane Collaboration search strategies were used. SELECTION CRITERIA: Studies for selection examined the effectiveness of interlock programs in a defined population. Studies were required to have a clear description of the program and outcomes evaluated, to have a comparison group and to provide interpretable data. DATA COLLECTION AND ANALYSIS: A total of 31 studies were found. Ten studies met the selection criteria. Three of these studies were eliminated from further analysis because they did not contain original data. A fourth study was eliminated due to methodologic weaknesses, leaving six studies for final review and analysis. Pooled analyses were not done because studies did not follow similar methods over comparable time periods. MAIN RESULTS: Five of the six studies found interlocks were effective in reducing DWI recidivism while the interlock was installed in the car. In the five studies demonstrating a significant effect, participants in the interlock programs were 15%-69% less likely than controls to be re-arrested for DWI. The only reported randomized, controlled trial demonstrated a 65% reduction in re-arrests for DWI in the interlock group, compared with the control group. CONCLUSIONS: Alcohol ignition interlock programs appear to be effective in reducing DWI recidivism during the time period when the interlock is installed in the car. Future studies should attempt to control for exposure (i.e., number of miles driven) and determine if certain sub-groups are most benefited by interlock programs.

Correlates of injury to women with partners enrolled in batterer treatment programs.

OBJECTIVES: This study examined correlates of injury in a cohort of women who were partners of men enrolled in batterer treatment programs. METHODS: Cross-sectional data of 670 pairs of battered women and their partners were analyzed. Prevalence rates of women's self-reports of injury due to their partners' abusive behavior were computed and compared by couples' demographic and behavioral characteristics. RESULTS: Men's use of severe tactics of abuse was the characteristic most significantly associated with injury (odds ratio = 15.47; 95% confidence interval = 9.02, 26.55). CONCLUSIONS: Our findings underscore the need to obtain information on the specific tactics used by couples to settle their disputes during universal screening by practitioners.

Prevalence of intimate partner abuse in women treated at community hospital emergency departments.

CONTEXT: The majority of prior studies examining intimate partner abuse in the emergency department (ED) setting have been conducted in large, urban tertiary care settings and may not reflect the experiences of women seen at community hospital EDs, which treat the majority of ED patients in the United States. OBJECTIVE: To determine the prevalence of intimate partner abuse among female patients presenting for treatment in community hospital EDs and describe their characteristics. DESIGN: An anonymous survey conducted from 1995 through 1997 inquiring about physical, sexual, and emotional abuse. SETTING: Eleven community EDs in Pennsylvania and California. PARTICIPANTS: All women aged 18 years or older who came to the ED during selected shifts. MAIN OUTCOME MEASURES: Reported acute trauma from abuse, past-year physical or sexual abuse, and lifetime physical or emotional abuse. RESULTS: Surveys were completed by 3455 (74%) of 4641 women seen. The prevalence of reported abuse by an intimate partner was 2.2% (95% confidence interval [CI], 1.7%-2.7%) for acute trauma from abuse, 14.4% (95% CI, 13.2%-15.6%) for past-year physical or sexual abuse, and 36.9% (95% CI, 35.3%-38.6%) for lifetime emotional or physical abuse. California had significantly higher reported rates of past-year physical or sexual abuse (17% vs 12%, P<.001) and lifetime abuse (44% vs 31%, P<.001) than Pennsylvania. Logistic regression modeling identified 4 risk factors for reported physical, sexual, or acute trauma from abuse within the past year: age, 18 to 39 years (odds ratio [OR], 2.2; 95% CI, 1.7-3.0); monthly income less than $1000 (OR, 1.7; 95% CI, 1.3-2.1); children younger than 18 years living in the home (OR, 2.0; 95% CI, 1.5-2.6); and ending a relationship within the past year (OR, 7.0; 95% CI, 5.5-8.9). CONCLUSION: If the prevalence of abuse in community hospitals throughout the United States is similar to the range of prevalence estimates found in this study, then heightened awareness of intimate partner abuse is warranted for patients presenting to the ED.

Incidence of dog bite injuries treated in emergency departments.

CONTEXT: Dog bites that result in injuries occur frequently, but how frequently dog bite injuries necessitate medical attention at a hospital or hospital admission is unknown. OBJECTIVE: To describe the incidence and characteristics of dog bite injuries treated in US emergency departments (EDs). DESIGN: Emergency department survey from the National Center for Health Statistics National Hospital Ambulatory Medical Care Survey for 1992 to 1994. PATIENTS: National probability sample of patients visiting EDs. MAIN OUTCOME MEASURE: Incidence of dog bites treated in EDs, defined as a cause of injury recorded as the E-code E906.0. RESULTS: The 3-year annualized, adjusted, and weighted estimate of new dog bite-related injury visits to US EDs was 333687, a rate of 12.9 per 10000 persons (95% confidence interval [CI], 10.5-15.4). This represents approximately 914 new dog bite injuries requiring ED visits per day. The median age of patients bitten was 15 years, with children, especially boys aged 5 to 9 years, having the highest incidence rate (60.7 per 10000 persons for boys aged 5 to 9 years). Children seen in EDs were more likely than older persons to be bitten on the face, neck, and head (73% vs 30%). We estimated that for each US dog bite fatality there are about 670 hospitalizations and 16000 ED visits. CONCLUSIONS: Dog bite injuries are an important source of injury in the US population, especially among children. Improved surveillance and prevention of dog bite-related injuries, particularly among children, are needed.

Victimization from physical violence in Pennsylvania: prevalence and health care use.

OBJECTIVES: To determine the prevalence of interpersonal physical violence (IPV) among Pennsylvania adults, to identify the personal characteristics of the victims, and to determine their health care use for resulting injuries. METHODS: Population-based data describing physical violence were obtained through a statewide telephone survey of 3,620 Pennsylvania adults selected from households by random-digit dialing in 1994. The prevalence and 95% confidence interval (95% CI) of victimization from IPV along with ED or other medical care facility use for IPV-related injuries were computed by several personal characteristics. Logistic regression was used to compare victims of IPV and their levels of health care use. RESULTS: The prevalence of reported victimization from IPV was 5.6% (95% CI = 4.9, 6.3). Significantly more victimization was reported by males, persons aged 18-29 years, those employed, and unmarried persons. The proportion of victims who reported to have gone to an ED or other medical care facility for IPV-related injury treatment was 12.9%. Significantly more persons with annual household incomes < $20,000 reported health care use for injuries resulting from IPV than did those with incomes of > or = $20,000 (OR = 3.98; 95% CI = 1.27, 12.48). Health care use for injuries was not found to be related to gender, age, race, employment, or marital status. CONCLUSIONS: This population-based study of health care use for IPV-related injuries found that victims of physical violence in Pennsylvania were not only young and unmarried men, but also employed. Health care use for resulting injuries was greater among persons with lower incomes.

Driving patterns and medical conditions in older women.

OBJECTIVES: To describe driving patterns (e.g., driving frequency) in older women drivers and to evaluate the impact of medical conditions and comorbidity on driving patterns. DESIGN: Cross-sectional examination of the association between medical conditions and driving patterns. SETTING: Population-based cohort from the Pittsburgh Center of the Study of Osteoporotic Fractures (SOF). PARTICIPANTS: A total of 1768 women aged 71 years or older. MAIN MEASUREMENTS: Driving information was obtained through a driving questionnaire, including driving status, weekly mileage, longest trip in the past year, etc. Data for demographics, lifestyle behavior, and medical conditions were collected through the SOF study. RESULTS: Among the participants, 1103 (62.3%) were current drivers, 337 (19.1%) had stopped driving, and 329 (18.6%) had never driven in their lifetime. The proportion reporting driving cessation and decline in driving amount increased with age. The prevalence of most medical conditions was higher among former drivers than in current or never drivers. Even after controlling for age and other demographic variables, fractures, heart disease, diabetes, self-reported poor vision or hearing, as well as comorbidity were found to be associated independently with decreased driving amount, including driving cessation, decline in mileage, and avoiding long trips. CONCLUSION: Both individual medical conditions and comorbidity influence driving patterns in older drivers. Because it is common for older people to have several medical conditions simultaneously, comorbidity might be a more comprehensive measure of medical impact on driving.

Dog bite incidence in the city of Pittsburgh: a capture-recapture approach.

OBJECTIVES: The purpose of this study was to estimate the number of dog bite injuries occurring in the city of Pittsburgh in 1993. METHODS: The capture-recapture method was used, along with long-linear modeling. Three sources were used to identify victims hospital reports, animal control reports, and police/victim reports. RESULTS: In 1993, 790 dog bites were reported. The capture-recapture method estimated that there were 1388 unreported dog bites, with an estimated incidence rate of 58.9 per 10,000. CONCLUSIONS: Dog bite is a common our preventable injury. To improve surveillance, the focus should be on educating the general public about the serious consequences of dog bite injuries.

A population-based study of fatal and nonfatal firearm-related injuries.

OBJECTIVE: To determine population-based firearm-related morbidity and mortality for Allegheny County, PA (population = 1.3 million), for the year 1994. METHODS: Fatalities were identified from a review of death certificates. To identify nonfatal cases, an active surveillance was conducted at all 24 acute care EDs in the county. The ED surveillance used 2 existing sources of case identification from each hospital to minimize undercount. RESULTS: Firearms were the leading cause of injury death to county residents, accounting for 155 deaths. The crude mortality rate from firearms was 11.7/100,000. Black males aged 15-19 years were most at risk for a firearm fatality (293/100,000). There were 514 nonfatal firearm injuries, producing a case fatality rate of 23%. The highest age-specific rate for nonfatal firearm-related injuries treated in the county EDs was observed for black males aged 15-19 years (2,245/100,000), which is 58 times higher than the firearm-related injury rate for the entire county population (38.7/100,000). CONCLUSION: Firearm-related injury and death are a significant public health problem in Allegheny County. Although the crude mortality rate from firearms in the county is lower than the reported national rate, the observed rate for nonfatal injuries in the black youth of this community is the highest firearm injury incidence rate ever reported. Local surveillance of firearm-related injuries, including nonfatal events, is needed to more accurately demonstrate the magnitude of this problem.

The potential of using billing data for emergency department injury surveillance.

OBJECTIVE: To determine the availability of and sample statewide ED injury information obtained from hospital billing data for the purpose of demonstrating the feasibility of information acquisition for subsequent data linkage. METHODS: A retrospective, database investigation was conducted to obtain data describing a statewide stratified sample of ED patients. The aim was to collect a computerized billing summary record for each injured ED patient seen at each sampled hospital over a 1-year period. All 215 Pennsylvania acute care hospitals in 1991 were eligible for sample selection. Data collection for the project was conducted in 1993. Participants included directors of hospital medical records and billing departments. RESULTS: Twenty-four hospitals contributed data sets from the original target goal of 31 strata. The final combined data set contained 187,404 records with injury diagnoses from approximately 616,000 ED patient visits, representing a 12% sample of all annual statewide ED visits. Age, sex, date of visit, and primary diagnosis fields were completed from the retrieved data > 99% of the time. More than two-thirds of the sampled records had a social security number, and total charges were recorded > 90% of the time. Other variables such as name and address were contained in < 50% of the records submitted. E-codes were usually not available. CONCLUSIONS: Retrospective compilation of multihospital ED billing data to create a statewide ED data sample-with the potential for injury research and probabilistic database linkage-can be accomplished; there are, however, important limitations.

Injury prevention: is there a role for out-of-hospital emergency medical services?

OBJECTIVE: To review the literature for options for integrating injury prevention into the role of out-of-hospital emergency medical services (EMS). DATA SOURCES: Computerized searches of the English-language literature from 1966 through 1994 were conducted using the MEDLINE and National Association of EMS Physicians (NAEMSP) databases. These were supplemented by hand searches of pertinent journals not indexed on MEDLINE or by NAEMSP and the reference lists of retrieved articles. Key words searched included emergency medical services, accident, injury, prevention, and safety. ARTICLE SELECTION: The review included all articles that described the experience of EMS organizations or individuals providing primary injury prevention (PIP) services or that proposed EMS PIP activities. SYNTHESIS: PIP EMS experiences and PIP activities proposed for EMS included: preventing injuries in EMS providers, serving as role models, identifying persons at risk for injury, providing prevention counseling, collecting injury data, surveying residences and institutions for injury risks and hazards, conducting educational programs and media campaigns, and advocating legislative changes that promote injury prevention. Few studies have evaluated the effectiveness of EMS PIP activities. CONCLUSION: As changes in the market compel health care systems to focus more on prevention, EMS organizations and individual providers may be assuming new injury prevention roles. Some EMS systems in many parts of the country have incorporated PIP into their work. It is necessary, however, to determine which PIP roles are effective and how they will be supported.

Identifying homes with inadequate smoke detector protection from residential fires in Pennsylvania.

With data from the Center for Disease Control's Behavioral Risk Factor Surveillance System, we estimated the prevalence of homes with inadequate smoke detector protection from residential fires in Pennsylvania and identified the characteristics of these homes in an effort to identify useful prevention strategies for Pennsylvania and other regions with similar characteristics. Homes with inadequate smoke detector protection from residential fires were defined on three levels. Eight percent (95% CL = 6.8, 8.6) of homes lacked any installed smoke detectors, 14% (95% CL = 12.8, 15) lacked smoke detectors installed on the same floor where they slept, and 28% (95% CL = 26.6, 29.6) were found to engage in unsafe smoke detector practices. The strong predictors of unsafe smoke detector practices determined from logistic regression included black, non-Hispanic homes (OR = 1.53), homes with annual household income of less than +20,000 (OR = 1.29), and those with no children younger than 5 years old (OR = 1.55). These findings should assist policy makers in planning residential fire prevention programs for Pennsylvania, which has yet to meet the Healthy People 2000 objective regarding fire prevention.

Injury surveillance: a statewide survey of emergency department data collection practices.

STUDY OBJECTIVE: To assess current emergency department data collection practices in Pennsylvania and determine whether existing data sources can be used as part of a statewide injury surveillance system. METHODS: Separate survey questionnaires requesting information on current ED patient data collection practices and attitudes were mailed to all directors of medical records, billing, and EDs in Pennsylvania (N = 212). RESULTS: Of the medical records department respondents, 92% indicated that ED registration data are retained in a computerized information system; 94% of respondents from billing departments reported that their ED patient registration system is integrated with an ED billing system. A total of 36% of EDs surveyed use a computerized ED patient logbook, and another 27% plan to begin a computerized log within 2 years. Dictation and transcription services that permit electronic retrieval of text are being used by 26% of EDs for patient medical records. CONCLUSION: Many elements for building a statewide ED injury surveillance system are in place in Pennsylvania, but they are as yet incomplete. Future studies should examine the feasibility of integrating existing ED data systems into statewide injury surveillance systems.

Firearm ownership and storage practices in Pennsylvania homes.

OBJECTIVE: To determine the household prevalence of firearms in Pennsylvania, and describe the storage practices for these weapons. DESIGN: A statewide telephone survey of 3,620 Pennsylvania adults selected from households by random digit dialing in 1994. MAIN OUTCOME MEASURES: Firearm ownership and storage practices were computed by household characteristics using logistic regression. RESULTS: The prevalence of firearm ownership was 37% (95% confidence interval = 35.4 to 38.6). Ownership of firearms was significantly higher for white residents, households with annual income of $20,000 or more, those in rural counties, and those with children and adolescents. Of the households with firearms, 23% contained a single firearm, the majority of which were handguns (40%) or rifles (40%); 76% had two or more firearms, with 57% reporting one handgun or more and 83% reporting one rifle or more. Storage of firearms in 72% of households involved two or more of these barriers: (1) taken apart; (2) trigger lock applied; (3) kept in a locked place; (4) unloaded; (5) no other ammunition; (6) locked ammunition; 6% stored at least one of their firearms with none of these barriers. The strongest predictor of storing a firearm with fewer than two protective barriers was households with no children or adolescents. CONCLUSIONS: Firearms are present in a large number of Pennsylvania homes. Many of these homes also contain children. To reduce the potential risks of firearms, optimal methods of storage of firearms in the home need to be determined.

Evaluation of the emergency department logbook for population-based surveillance of firearm-related injury.

STUDY OBJECTIVE: To evaluate existing emergency department logbooks as a source of population-based data on firearm-related injuries. METHODS: We examined the logbooks of the 24 acute care and specialty-hospital EDs in Allegheny County, Pennsylvania, to determine the number and type of data variables each contained and the completeness of reporting of each variable for selected firearm-related cases. The amount of missing data for certain variables was determined and the cause for the missing data described. RESULTS: Logbooks from 18 of the 24 eligible hospitals were reviewed. We identified 785 cases of firearm-related injury recorded between January 1, 1992, and December 31, 1993. Of the variables we selected for analysis, only date (100%), chief complaint or diagnosis (100%), name (98%), and time of admission (97%) were consistently documented. In 37% of cases the patient's county of residence could not be determined. Similarly incomplete data were found for body part injured (31%), race (28%), age (26%), sex (22%), and mode of arrival (21%). The factor most responsible for the high percentage of incomplete data was the considerable variation in the data elements contained in the different hospitals' logbooks. CONCLUSION: Missing data resulting from inconsistencies in the variables contained in different EDs' logbooks and errors of omission prevent ED logbooks, in their current state, from providing population-based data for surveillance of firearm-related injury. Standardization of such variables in ED logbooks would yield a more useful source of information for injury and disease surveillance. In lieu of standardized logbooks, multiple sources of data are necessary to establish a more comprehensive and useful system of surveillance of firearm-related injury.