Evaluating the impact of a street barrier on urban crime.

OBJECTIVES: Violence is a major urban public health problem in the United States. The impact of a physical barrier placed across a street in a public housing project to prevent street violence and drug activity was evaluated. METHODS: Hartford Police Department data on violent and drug related crime incidence within the housing project containing the barrier were analyzed by use of a computerized geographic information system. RESULTS: Violent crime decreased 33% on the intervention street during the 15 month period after erection of the barrier, compared with the 15 month period before erection of the barrier, but there was no change in drug related crime. On adjoining streets and surrounding blocks, violent crime decreased 30%-50% but drug related crimes roughly doubled. A non-adjacent area of the housing project and the entire city experienced 26% and 15% decreases in violent crimes, and 414% and 25% increases in drug crimes, respectively. CONCLUSIONS: The barrier decreased violent crime but displaced drug crimes to surrounding areas of the housing project. These results have important implications for other cities that have erected or are considering erecting similar barriers.

Injury prevention training in pediatric residency programs.

OBJECTIVES: To quantify the content and setting of injury prevention training provided to pediatric residents and to identify aspects of residency programs associated with this training. DESIGN: Mail survey. SETTING: US pediatric residency programs. PARTICIPANTS: Residency program directors. MAIN OUTCOME MEASURES: Number of programs stating that they train residents in various injury prevention topics, and the setting of this training. RESULTS: More than 80% of programs addressed 92% of topics surveyed dealing with disease prevention, but only 59% of topics dealing with injury prevention. Injury prevention topics covered less frequently included smoke detector use and swimming pool, sports, and firearm safety. Programs used continuity clinics most often to teach injury prevention. No specific program characteristic was associated with the content or setting of injury prevention training. Residency programs located in states in which house fires, drowning, or firearm injuries are the leading causes of death were not more likely to offer prevention training on these topics. CONCLUSIONS: Injury prevention is less frequently taught than disease prevention in pediatric residency training. Injury prevention is most often taught in continuity clinics, the setting most consistent with ongoing primary patient care. A gap exists between the leading causes of injury death and injury prevention topics taught to pediatric residents. Residency programs must better recognize and adapt to the epidemiology of trauma in their communities, better enabling new pediatricians to meet their patient's needs.

Evaluation of the effectiveness of a pavement stencil in promoting safe behavior among elementary school children boarding school buses.

OBJECTIVE: The majority of school bus-related fatalities among children attending elementary school in the United States occur as children board or alight from buses. Injuries occur during boarding when children enter the street and are struck by buses or other vehicles. This study evaluated the effectiveness of a stencil in the shape of a school bus applied to the pavement at a bus stop in improving safe behaviors at bus stops. Specifically, we assessed the frequency of children running toward the bus as it approached or entered the street. METHODS: Elementary school bus stops with similar roadways, traffic profiles, and numbers of children boarding participated in the study. Stops were randomly assigned to an intervention group, in which children were instructed to remain within a safe area during boarding that was demarcated by a pavement stencil, or an education-only group, in which the safe area was demarcated by some existing environmental feature. Both groups received education about safe boarding procedures. Observers rated behavior at each stop daily for 5 consecutive weeks. Data were analyzed as bivariate odds of any unsafe behavior in the education-only group. RESULTS: One hundred forty-five observations from seven bus stops with stencils and 174 daily observations from six education-only stops were completed. Observations of children in the education-only group were twice as likely to show unsafe behavior while waiting (odds ratio [OR], 2.1; 95% confidence interval [CI], 1.3-3.6) and during boarding (OR, 2.1; 95% CI, 1.2-3.9). ORs were significantly higher in the education-only group for boys, girls, and children in grades 4 through 6. When no adult was present, there was a greater likelihood of unsafe behavior among all children in the education-only group while waiting (OR, 16.1; 95% CI, 3.9-72.4) and during boarding (OR, 15.0; 95% CI, 3.2-81.4). The presence of an adult at the stop did not have an independent effect on behavior. Children at education-only stops located on roadways with high traffic volume were more likely to engage in unsafe behavior while waiting (OR, 8.0; 95% CI, 3.8-17.3) and during boarding (OR, 4.9; 95% CI, 2.2-11.0). No differences were observed during boarding between stencil and education-only groups when 10 or more children were at the stops. CONCLUSION: The pavement stencil, when accompanied by education about safe boarding, may represent a cost-effective approach to reducing unsafe behavior at bus stops by children of elementary school age.

A population-based study of severe firearm injury among children and youth.

BACKGROUND: Firearm injuries are a major cause of pediatric mortality and morbidity in the United States. To date, population-based studies describe the epidemiology of firearm-related deaths; however, the patterns of severe, nonfatal pediatric firearm-related injuries are not as well defined. OBJECTIVES: To determine the epidemiology of severe firearm-related deaths and injuries among a statewide population of children and youth ages birth to 19 years. METHODS: Demographic, geographic, and cost data were analyzed from Connecticut death certificates for 1988 through 1992 and hospital discharge data for 1986 through 1990 for firearm-related unintentional, self-inflicted, and assaultive injury among children and youth ages birth to 19 years. RESULTS: There were 219 firearm deaths: 68% homicides, 25% suicides, 6% unintentional, and 1% of undetermined intent, resulting in an annual age-specific death rate of 6.6 per 100,000 persons. There were 533 hospitalizations for gunshot wounds (16 per 100,000); 41% were assaults, 1% suicide attempts, 39% unintentional gunshot wounds, 1% legal interventions, and 18% of undetermined intent. More than 80% of deaths from gunshot wounds and hospitalizations occurred among 15- to 19-year-old males, most occurring in Connecticut's five largest cities. Most firearm homicides occurred among urban residents; most firearm suicides occurred among nonurban residents; and unintentional shootings were evenly distributed between urban and nonurban residents. The total cost of firearm-related hospitalizations averaged $864,000 per year. CONCLUSIONS: Firearms are a major cause of mortality and morbidity of Connecticut children and youth, exceeded only by motor vehicles as a cause of death among those 1 to 19 years of age. Handguns were responsible for a disproportionate amount of trauma compared with other firearm types. The epidemiology of pediatric gunshot injuries requires a range of strategies for prevention. Physicians caring for families with children must include firearm injury prevention counseling as a routine part of anticipatory guidance.

Pediatric injury prevention.

Pediatric injuries are the leading cause of childhood death and disability and are responsible for more childhood deaths than all other diseases combined. The panel summarized the principles of pediatric injury prevention and reviewed the incidence, epidemiology, and prevention of six common pediatric injuries.

Misclassification of childhood homicide on death certificates.

Suspect classification of homicide deaths of Connecticut residents under 20 years of age was noted for 29 percent of cases examined. Misclassification was attributed to incomplete or erroneous information recorded on the death certificates, rather than errors in the designation of ICD-9 homicide codes. The results have important implications in the interpretation of vital statistics when homicide is listed as the cause of death and underscore the value of record linkage systems.