Showing papers in "Injury Prevention in 1997"

Epidemiology of bicycle injuries and risk factors for serious injury.

Abstract: OBJECTIVE: To determine the risk factors for serious injury to bicyclists, aside from helmet use. DESIGN: Prospective case-control study. SETTING: Seven Seattle area hospital emergency departments and two county medical examiner's offices. PATIENTS: Individuals treated in the emergency department or dying from bicycle related injuries. MEASUREMENTS: Information collected from injured bicyclists or their parents by questionnaire on circumstances of the crash; abstract of medical records for injury data. Serious injury defined as an injury severity score > 8. ANALYSIS: Odd ratios computed using the maximum likelihood method, and adjusted using unconditional logistic regression. RESULTS: There were 3854 injured cyclists in the three year period; 3390 (88%) completed questionnaires were returned 51% wore helmets at the time of crash. Only 22.3% of patients had head injuries and 34% had facial injuries. Risk of serious injury was increased by collision with a motor vehicle (odds ratio (OR) = 4.6), self reported speed > 15 mph (OR = 1.2), young age ( 39 years (OR = 2.1 and 2.2 respectively, compared with adults 20-39 years). Risk for serious injury was not affected by helmet use (OR = 0.9). Risk of neck injury was increased in those struck by motor vehicles (OR = 4.0), hospitalized for any injury (OR = 2.0), and those who died (OR = 15.1), but neck injury was not affected by helmet use. CONCLUSIONS: Prevention of serious bicycle injuries cannot be accomplished through helmet use alone, and may require separation of cyclists from motor vehicles, and delaying cycling until children are developmentally ready.

Fatal and non-fatal farm injuries to children and adolescents in the United States, 1990-3.

Abstract: OBJECTIVE: Examine the current magnitude of the injury problem to children and adolescents on farms, and to compare these data to that from 1978-83. DATA SOURCES: US National Center for Health Statistics Mortality Multiple Cause of Death Tapes for the years 1991-3, and the US Consumer Product Safety Commission National Electronic Injury Surveillance System for data on emergency department visits for 1990-3. SUBJECTS: Children and adolescents 19 years and younger injured on farms. RESULTS: There were an average of 104 deaths per year due to injuries occurring on farms. The rate of 8.0 deaths per 100,000 child farm residents is 39% lower than in 1979-81. More of the deaths occurred in hospital than previously. There were an average of 22,288 emergency department treated injuries per year. The rate of 1717 injuries per 100,000 child farm residents is 10.7% higher than 1979-83. Males were injured more frequently than females. Tractors accounted for 20.9% of all injuries, followed by horses (8.4%), all terrain vehicles and minibikes (8.0%), and farm wagons (7.7%). CONCLUSIONS: Farm injuries continue to be a major problem to children living on farms. While improved medical care may have contributed to the reduction in mortality, the continued high rate of injuries warrants study of a variety of intervention strategies to reduce the injury toll. There is also a need for ongoing injury surveillance to provide accurate data on the farm injury problem.

A review of risk factors for child pedestrian injuries: are they modifiable?

Abstract: PURPOSE: To identify modifiable risk factors for child pedestrian injuries. DATA SOURCES: (1) MEDLINE search from 1985 to 1995; search term used was traffic accidents; (2) review of reference lists from retrieved articles and books; (3) review of reference lists from three systematic reviews on childhood injuries and (4) consultation with 'key informants'. STUDY SELECTION: All studies that examined the risk factors for child pedestrian injuries were targeted for retrieval. Seventy potentially relevant articles were identified using article titles, and, when available, abstracts. Of the 70 retrieved articles, 44 were later assessed as being relevant. QUALITY ASSESSMENT: Articles were classified on the basis of study design as being either descriptive (hypothesis generating) (26) or analytical (hypothesis testing) (18) studies. Consensus was used for difficult to classify articles. DATA EXTRACTION: Variables judged to be risk factors for child pedestrian injuries were extracted by one author. DATA SYNTHESIS: A qualitative summary of the information extracted from relevant articles is presented in tabular form. RESULTS: Risk factors for child pedestrian injuries were classified as: (1) child, (2) social and cultural, (3) physical environment, and (4) driver. Risk factors within each classification are summarized and discussed.

Factors affecting the severity of motor vehicle traffic crashes involving young drivers in Ontario.

Abstract: OBJECTIVES: To assess the factors affecting the severity of motor vehicles traffic crashes involving young drivers in Ontario. POPULATION: Ontario young drivers, aged 16 to 20, involved in traffic crashes resulting in injury, between 1 January 1988 and 31 December 1993, on public roads in Ontario. METHODS: Population based case-control study. Cases were fatal injury, major injury, and minor injury crashes involving young drivers. Controls were minimal injury crashes involving young drivers. Cases and controls were obtained retrospectively from the Canadian Traffic Accident Information Databank. Unconditional logistic regression was used for data analysis. RESULTS: Factors significantly increasing the risk of fatal injury crashes include: drinking and driving (odds ratio (OR) 2.3), impairment by alcohol (OR 4.8), exceeding speed limits (OR 2.8), not using seat belts (OR 4.7), full ejection from vehicle (OR 21.3), intersection without traffic control (OR 2.2), bridge or tunnel (OR 4.1), road with speed limit 70-90 km/hour (OR 5.6) or 100 km/hour (OR 5.4), bad weather (OR 1.6), head-on collision (OR 80.0), and overtaking (OR 1.9). Results of the same model applied to major and minor injury crashes demonstrated consistent but weaker associations with decreasing levels of crash severity. CONCLUSIONS: A casual relationship between crash severity and the risk factors listed above was proposed. Risk factors recommended for preventive intervention include: alcohol consumption, speeding, and use of seat belts. Head-on collisions are of primary concern.

An international study of the exposure of children to traffic.

Abstract: OBJECTIVES: To examine the extent of international differences in children's exposure to traffic as pedestrians or bicyclists. DESIGN: Children's travel patterns were surveyed using a parent-child administered questionnaire. Children were sampled via primary schools, using a probability cluster sampling design. SETTING: Six cities in five countries: Melbourne and Perth (Australia), Montreal (Canada), Auckland (New Zealand), Umea (Sweden), and Baltimore (USA). SUBJECTS: Children aged 6 and 9 years. MAIN OUTCOME MEASURES: Modes of travel on the school-home journey, total daily time spent walking, and the average daily number of roads crossed. MAIN FINDINGS: Responses were obtained from the parents of 13423 children. There are distinct patterns of children's travel in the six cities studied. Children's travel in the three Australasian cities, Melbourne, Perth and Auckland, is characterised by high car use, low levels of bicycling, and a steep decline in walking with increasing car ownership. In these cities, over a third of the children sampled spent less than five minutes walking per day. In Montreal, walking and public transport were the most common modes of travel. In Umea, walking and bicycling predominated, with very low use of motorised transport. In comparison with children in the Australasian and North American cities, children in Umea spend more time walking, with 87% of children walking for more than five minutes per day. CONCLUSIONS: There are large international differences in the extent to which children walk and cycle. These findings would suggest that differences in 'exposure to risk' may be an important contributor to international differences in pedestrian injury rates. There are also substantial differences in pedestrian exposure to risk by levels of car ownership-differences that may explain socioeconomic differentials in pedestrian injury rates.

The PRECEDE-PROCEED model: application to planning a child pedestrian injury prevention program.

Abstract: OBJECTIVES: The objectives were first, to modify the PRECEDE-PROCEED model and to use it is as a basis for planning a three year intervention trial that aims to reduce injury to child pedestrians. A second objective was to assess the suitability of this process for planning such a relatively complex program. SETTING: The project was carried out in 47 primary schools in three local government areas, in the Perth metropolitan area. METHODS: The program was developed, based on extensive needs assessment incorporating formative evaluations. Epidemiological, psychosocial, environmental, educational, and demographic information was gathered, organised, and prioritised. The PRECEDE-PROCEED model was used to identify the relevant behavioural and environmental risk factors associated with child pedestrian injuries in the target areas. Modifiable causes of those behavioural and environmental factors were delineated. A description of how the model facilitated the development of program objectives and subobjectives which were linked to strategy objectives, and strategies is provided. RESULTS: The process used to plan the child pedestrian injury prevention program ensured that a critical assessment was undertaken of all the relevant epidemiological, behavioural, and environmental information. The gathering, organising, and prioritising of the information was facilitated by the process. CONCLUSIONS: The use of a model such as PRECEDE-PROCEED can enhance the development of a child injury prevention program. In particular, the process can facilitate the identification of appropriate objectives which in turn facilitates the development of suitable interventions and evaluation methods.

Sibling power: influence of older siblings' persuasive appeals on younger siblings' judgements about risk taking behaviours.

Abstract: OBJECTIVES: Although many injuries happen when school age children are away from home and in the company of other children, we know surprisingly little about interpersonal influences on children's risk taking decisions. The aim of the present study was to examine the influence of older siblings' persuasive appeals on young children's decisions about engaging in behaviours that could threaten their physical safety. METHODS: Forty same sex sibling pairs participated. Children were shown drawings of play scenes (bicycling, river crossing, and sledding), with each depicting lower and higher risk paths of travel. Children of 8 years made initial decisions as to which paths they would take. Subsequently, their older sibling acted as a confederate and tried to persuade them to change their decisions. RESULTS: After the appeals of older siblings, younger children significantly shifted their decisions: choices of less risky paths replaced the initial selection of more risky paths, and vice versa. A positive sibling relationship was predictive of younger siblings' decision changes. Boys and girls were equally effective in persuasion but they did so using different types of arguments, with boys communicating primarily appeals to fun and girls emphasizing appeals to safety. CONCLUSIONS: These findings highlight the effect that older siblings can have on risk taking decisions of younger siblings. Accordingly, they document the importance of considering the interpersonal context of risk taking when designing interventions to reduce injuries among elementary schoolchildren.

Home drowning among preschool age Mexican children.

Abstract: OBJECTIVES: To estimate the risk of drowning by different bodies of water in and near the home for children aged 1 to 4 years. SETTING: The Metropolitan Area of Guadalajara, Mexico. METHODS: A population case-control study. Cases (n=33) were children 1 to 4 years old who drowned at their home; controls (n=200) were a random sample of the general population. RESULTS: The risk of drowning for children whose parents reported having a water well at home was almost seven times that of children in homes without a water well (adjusted odds ratio (OR)=6.8, 95% confidence interval (CI)=2.2 to 20.5). Risk ratio estimates for other bodies of water were: swimming pools (OR=5.8, 95% CI=0.9 to 37.5), water barrel (OR=2.4, 95% CI=1.0 to 5.6), underground cistern (OR=2.1, 95% CI=0.8 to 5.2), and a basin front (courtyard pool to store water) of 35 or more litres (OR=1.8, 95% CI=0.8 to 4.4). CONCLUSION: Drowning at home is frequent in the Metropolitan Area of Guadalajara, but the causes are different from those reported in developed countries. Accordingly, the preventive strategies must also be different.

A descriptive analysis of children's playground injuries in the United States 1990-4.

Abstract: OBJECTIVES: To review playground injury statistics over a five year period in order to develop an awareness of how and where children in the United States are being injured. METHODS: All data are based on the United States Consumer Product Safety Commission's National Electronic Injury Surveillance System (NEISS) for playground related injuries during 1990-4. The surveillance data includes injuries recorded in more than 90 hospital emergency departments located throughout the United States. RESULTS: Each year there are roughly 211,000 preschool or elementary school-children in the United States who receive emergency department care for injuries associated with playground equipment. On average, 17 of these cases result in death. 70% of all injuries occur on public playgrounds, with nearly one third classified as severe. Swings, climbers, and slides are the pieces of playground equipment associated with 88% of all NEISS reported injuries. Falls to the surface are responsible for 70%. CONCLUSIONS: NEISS playground injury statistics contribute to our understanding of playground injuries. By identifying where and how children are injured, suggestions can be made in an attempt to make playgrounds safer.

Socioeconomic and racial/ethnic factors affecting non-fatal medically attended injury rates in US children.

Abstract: OBJECTIVE: Using a representative survey of US children, the purpose was to evaluate separate effects of socioeconomic and racial/ethnic factors, including access to care, on medically attended non-fatal injury rates. METHODS: Multivariate linear regression models were used to determine associations between injuries and health care coverage (insurance or Medicaid), having a place to go for care, race/ethnicity, maternal education, number of adults and children in the household, poverty, and urbanicity. The 1988 Child Health Supplement to the National Health Interview Survey included questions on medically attended injuries, and their cause, location, and effects on the child. Injury categories included total, consequential, occurrence at home or school, falls, and being struck or cut. RESULTS: Lack of health care coverage was consistently associated with lower medically attended injury rates in non-Hispanic blacks or whites and Mexican-Americans, but affected total rates for each group differently due to unequal distribution of health care coverage. Injuries occurred about 40% more frequently to children and adolescents living in single adult households compared with two adult homes for all injury categories except for injuries occurring at school. CONCLUSIONS: Preventive interventions targeted to specific populations based on assumptions that poverty, lack of education, or minority status result in greater risks for injuries require a closer look. Efficient targeting should address underlying factors such as differences in exposures and environments associated with single adult homes or recreational activities. Data sources used to target high risk populations for interventions need to address bias due to access to care.

Infant walker use, injuries, and motor development.

Abstract: OBJECTIVES: To study the impact of infant walker use on motor development and injuries. POPULATION: One hundred and eighty five parents or primary care givers who attended a Singapore government polyclinic from September 1993 to February 1994, with their infants between 7 to 10 months, for a developmental assessment session. SETTING: A government polyclinic in Singapore. METHODS: The parent or primary care giver answered questions pertaining to infant walker use and injuries attributed to its use. Each infant was then given the Singapore modified version of the Denver Developmental Screening Test (DDST-S), along with a full clinical examination; both testers were blinded to walker use. RESULTS: One hundred and sixty seven (90%) of 185 infants used walkers regularly, and 21 (12.5%) of the users had one or more injuries. Most injuries were minor, such as bruises and swellings on the head, forehead, face, and cheeks. None of the children who did not use walkers showed any abnormal DDST-S results whereas 18 (10.8%) of the 167 walker users had either abnormal or questionable DDST-S results. CONCLUSIONS: 12.5% of walker users had one or more injuries and walker use may also delay the child's motor development. These findings will help the physician or nurse in primary care settings to advise parents about the potential hazards of walker use.

High levels of incorrect use of car seat belts and child restraints in Fife--an important and under-recognised road safety issue.

Abstract: OBJECTIVE: To pilot data collection instruments and to make a preliminary estimate of the level of incorrect use of car seat belts and child restraints in Fife, Scotland. DESIGN: Cross sectional survey of cars containing adults and children at a number of public sites across Fife in 1995 to assess use of car occupant restraints. Trained road safety officers assessed whether seat restraints were appropriate for the age of the passengers and whether restraints were used correctly. These assessments were based on standards published by the Child Accident Prevention Trust. PARTICIPANTS: The survey gathered data from 596 occupants in 180 cars: 327 adults and 269 children. Ten per cent of drivers who were approached refused to participate. Car occupant restraint was assessed in 180 drivers, 151 front seat passengers, and 265 rear seat passengers. MAIN RESULTS: Three hundred and sixty one occupants wore seat belts, 68 were restrained by a seat belt and booster cushion, 63 in toddler seats, 25 in two way seats, and 18 in rear facing infant carriers. Ninety seven per cent of drivers, 95% of front seat passengers, and 77% of rear seat passengers were restrained. However, in 98 (52%) vehicles at least one passenger was restrained by a device that was used incorrectly. Seven per cent of adults and 28% of children were secured incorrectly. The commonest errors were loose seat belts and restraint devices not adequately secured to the seat. Rates of incorrect use were highest in child seat restraints, reaching 60% with two way seats and 44% with rear facing infant seats. CONCLUSIONS: The incorrect use of car occupant restraints is an under-recognised problem, both by health professionals, and the general public. Incorrect use has been shown to reduce the effectiveness of restraints, can itself result in injury, and is likely to be an important factor in child passenger injuries. The correct use of car seat restraints merits greater attention in strategies aiming to reduce road traffic casualties. Areas of intervention that could be considered include raising public awareness of this problem, improving information and instruction given to those who purchase child restraints, and encouraging increased collaboration between manufacturers of cars and child restraints, in considering safety issues.

Bike speed measurements in a recreational population: validity of self reported speed.

Abstract: OBJECTIVE: Speed at the time of a bicycle crash is an important determinant of the amount of energy transmitted. Controlling for speed is thus important in the evaluation of outcomes and effectiveness of intervention strategies. This study was conducted to evaluate the accuracy of self reported speed in a population of recreational cyclists. METHODS: Children's and adults' bicycle speeds were measured with a radar gun as they rode along a closed road at weekend recreational events. Cyclists were then stopped and asked to estimate their speed. Measured speed, cyclist's estimate of their speed, age, and sex were documented. Parents were also asked to estimate their child's speed. RESULTS: One hundred and fifty two cyclists from 4 to 80 years of age participated. Seventy per cent were children 13 years of age or younger. The mean (SD) speed of this group was 8.9 (2.5) mph. Cyclists age 14 and older were traveling at a mean speed of 9.7 (2.87) mph. Estimated speeds were significantly higher than measured speeds for this older group, but there was no significant difference between mean measured and estimated speeds for the younger riders. There was also no significant difference between measured and estimated speed for males and females. Parents estimated their child's speed quite accurately. CONCLUSIONS: Self reported speeds for children were in close agreement with measured speeds and, thus, are sufficiently accurate to be a useful measure of crash severity in evaluating helmet effectiveness.

Injury mortality among children and teenagers in New Zealand compared with the United States of America.

Abstract: OBJECTIVES: New Zealand (NZ) has an unenviable track record in childhood injury mortality. We sought to describe this burden and to compare it with the United States of America (USA), with a view to taking the first step in identifying potential areas in which NZ might benefit from injury control as practiced in the USA. METHODS: We identified all children and teenagers who had died of injury for the period 1984-93 from the NZ Health Information Service mortality data files. We compared their rates of injury with previously published rates for USA. RESULTS: The age specific rates follow a J shaped distribution, with high rates in the first year of life followed by a decline to the lowest rate, among 5-9 year olds, a marginally higher rate among 10-14 year olds, and a dramatic rise among those in the 15-19 age group. The specific causes of death vary considerably by age group. NZ's overall rate of child and adolescent injury is not substantially different from that of the USA, but marked differences are apparent when examining cause specific rates. CONCLUSIONS: In terms of the theoretical potential to reduce the total injury mortality rate, priority must be given to 15-19 year olds who account for 61% of all NZ injury deaths. Priorities for this age group are: motor vehicle traffic crashes (especially those involving occupants and motorcyclists), and suicide. Among the children, priorities are: pedestrian and occupant deaths, and drownings. Among infants, the priority is suffocation.

Non-fatal asphyxiation and foreign body ingestion in children 0-14 years.

Abstract: OBJECTIVES: To examine the frequency and nature of non-fatal asphyxiation and foreign body ingestion injuries among children in the state of Victoria, Australia, and to identify possible areas for prevention. METHODS: For children under 15 years, all Victorian public hospital admissions, July 1987 to June 1995, due to asphyxiation or 'foreign body entering through other orifice' (which includes ingestions), were reviewed. Emergency department presentations due to asphyxiation and foreign body ingestion provided information on circumstances of, and the type of foreign bodies involved in the injuries. RESULTS: The childhood average annual admission rate for asphyxiation was 15.1 per 100,000. Food related asphyxiation peaked in infants under 1 year, and declined to low levels by 3 years. The main foods involved were nuts, carrot, apple, and candy. The rate of non-food related asphyxiation was relatively constant to 3 years of age and then declined by 6 years. Mechanical suffocation was less common. The annual admission rate for 'foreign body entering through other orifice' was 31.7 per 100,000. These injuries peaked in 2-3 year olds then gradually declined. About 80% of these foreign body admissions were ingestions, with coins being the major object ingested. Admission rates for these causes remained constant over the eight years. Asphyxiation resulted in a higher proportion admitted and longer hospital stays. CONCLUSIONS: Prevention of suffocation and strangulation needs to focus on a safe sleeping environment and avoidance of ropes and cords, while foreign body asphyxiation and ingestion needs a focus on education of parents and child carers regarding age, appropriate food, risk of play with coins, and other small items. Legislation for toy small parts could be extended to those used by children up to the age of 5 years, and to other products marketed for children. Design changes and warning labels also have a place in prevention.

Childhood injury deaths: rural and urban differences, Colorado 1980-8.

Abstract: OBJECTIVES: The purpose of this study is to describe and compare the distribution of injury deaths among rural and urban Colorado children that occurred between 1980-8. METHOD: Death certificates coded E800-E969 were obtained for children who were 0-14 years of age at death between 1980-8 and who were Colorado residents. Average annual rates were computed for rural and urban children, separately by gender. The 1980 census was used to compute rates. Rate ratios were calculated to summarize information related to specific external causes of deaths, contrasting rural and urban children. These differences were evaluated using Z tests. RESULTS: Statistically significant elevated risks were found along rural children for motor vehicle injury deaths. Firearms were involved more often in rural deaths among unintentional injury deaths of children older than 4 years of age, and among homicide related deaths of children 5 years and younger. All rural children who committed suicide used a firearm. CONCLUSIONS: Comparisons between rural and urban injury deaths provide important information that can be used to guide prevention strategies. For example, in Colorado, a child restraint law, passed in 1984, covered children under 4 years of age or under 40 pounds. It was not until 1995, however, that legislation was passed requiring restraint of children 5-16 years of age. Traditionally, rural residents are slower to accept new ideas and to alter current practices than urban residents.

Injuries among children in home and out-of-home care.

Abstract: As the number of children receiving care in out-of-home settings increases in the United States, the risk of injury in such settings has become the subject of intense research. OBJECTIVES: This study examined the relative safety of out-of-home care compared with care in a child's own home. METHODS: This community based prospective cohort study of 656 families in three adjacent counties in the Piedmont region of North Carolina characterizes the patterns and rates of injuries among children less than 5 years of age in three child care settings, home care (HC), center based care (CBC), and other out-of-home care (OOHC). Information about minor and severe injuries was obtained from parents using monthly telephone interviews over a one year period. Statistical modeling designed to handle unbalanced data with correlated observations was used as the primary tool for analysis. RESULTS: Rate of minor injuries was highest in CBC, followed by HC, and then OOHC. However, these differences for OOHC may have been due to reporting biases and errors in rate estimates. There were no significant differences in severe injury rates among the three settings. CONCLUSIONS: The risk of serious injury among children under 5 in CBC is not different from that of children in HC or OOHC despite the fact that the risk of minor injury is higher.

Personality characteristics of the child accident repeater

Abstract: From 8874 boys and girls aged 4 to 18 in Berkeley-Oakland, California, 684 were selected to represent high-, intermediateor low-accident-liability children, based on records of medically attended injuries. Using data from intensive interviews with mothers supplemented with school records, we found a statistically significant relation between accident liability and indexes of extraversion, daring, roughhousing, and other traits tending to expose children to hazards. Similar relations held for traits such as poor discipline, aggressiveness toward peers, and, for girls, attention-seeking, which compete with the child's ability to cope with hazards. Other traits that may impair ability to cope with hazards were also found to be related to accident liability (impulsivity, carelessness, and unreliability) as were several indexes denoting maladjustment.

Characteristics and outcomes of self inflicted pediatric injuries: the role of method of suicide attempt.

Abstract: OBJECTIVE: To examine the epidemiologic characteristics and clinical outcomes of self inflicted pediatric injuries in relation to the method of suicide attempt. METHODS: Using data from the National Pediatric Trauma Registry Phase II, a comparative analysis was conducted for children under 15 years of age who were admitted from 1 October 1988 through 30 April 1996 because of self inflicted injury by firearm (n = 28), hanging (n = 38), or jumping from heights (n = 21). RESULTS: Of the 87 cases under study, 90% occurred at home, and 86% occurred between noon and midnight, with a peak in early evening (between 6 pm and 7 pm)-More than one quarter (29%) had preexisting mental disorders, such as disturbance of conduct and depression. Toxicological tests were conducted on admission on 40 (46%) of the patients; 20% tested positive for alcohol or other illicit drugs. The method of suicide attempt was associated with gender and age of the patients: 75% of the firearm cases and 82% of the hanging cases were boys compared with 29% of the jumping cases (p

Swimming pool immersion accidents: an analysis from the Brisbane Drowning Study. 1977.

Abstract: An analysis of a consecutive series of 66 swimming pool immersion accidents is presented; 74% of these occurred in in-ground swimming pools. The estimated accident rate per pool is fives times greater for in-ground pools compared with above-ground pools, where pools are inadequately fenced. Backyard swimming pools account for 74% of pool accidents. Motel and caravan park pools account for 9% of childhood immersion accidents, but the survival rate (17%) is very low. Fifty per cent of pool accidents occur in the family's own backyard pool, and 13.6% in a neighbour's pool; in the latter the survival rate is still low at only 33%. In only one of the 66 cases was there an adequate safety fence; in 76% of cases there was no fence or barrier whatsoever. Tables of swimming pool accidents by age, season, site, and outcome are presented.

Threats to measurement validity in self reported data can be overcome.

Abstract: The article by Evans and colleagues in this issue of the journal (p29) is exemplary of the strengths and weaknesses of descriptive data. The authors succeeded in gathering a broad range of information (for example, knowledge, attitudes, and self reported behaviors on several injury risks) from a diverse group of parents (for example, low income, high income, employed, unemployed) using a pilot tested mailed survey based on previously published data collection instruments. The authors failed, however, in attaining only a moderate response rate despite two attempts at follow up (for example, 67% and 58% from most and less affluent families, respectively). In addition, they measured socioeconomic status indirectly through residence; rather, each respondent should have been asked to provide household income for cross verification. Further, the authors failed to include a neutral option (for example, 'neither agree nor disagree') among the forced choice selections (for example, 'strongly agree', 'agree', 'disagree', 'strongly disagree'), resulting in strikingly reduced response variation. Despite these shortcomings, this research is decidedly representative of the published survey findings of parental knowledge, attitudes, and self reported behaviors concerning childhood injury prevention. Criticisms of the measurement validity of self reported data on childhood injury prevention practices abound, as aptly articulated in my colleague's opinion. Validity is defined as the 'adequacy with which the method of measurement does its job-how well does it measure the characteristic that the investigator actually wants to measure?\" The measurement validity of mailed surveys and telephone interviews are often viewed as suspect as they may elicit socially desirable responses resulting in over-reporting of safety behaviors and underreporting of risky behaviors by parents and caregivers. Although self report methods admittedly may not be as rigorous as well designed observational surveys in measuring behavior, they have several advantages. Data are much less expensive to amass, more expedient to collect, and provide information otherwise not accessible directly through observationnamely, respondent's knowledge, attitudes, and opinions.2 Further, this research method has been used extensively in the injury prevention field, specifically3-5 and in the public health field, generally (for example, National Health Interveiw Survey, Behavioral Risk Factor Surveillance Survey, US Census). Research findings; in turn, have underpinned the design and implementation of countless safety interventions. To discount the validity of self reported data would be tantamount to positing that injury prevention practitioners have been intervening for years in the safety habits of families on, at best, shaky evidence. I believe that self report surveys can yield valid data, providing the researcher acknowledges possible sources of error, and if certain precautions are taken.

Socioeconomic status and the prevention of child home injuries: a survey of parents of preschool children.

Abstract: OBJECTIVES: To examine the effect of socioeconomic status on the attitudes parents of preschool children towards child home safety issues and practice of home safety measures. SETTING: A community based study in the Lanarkshire Health Board area, a mixed urban-rural setting in central Scotland. METHODS: A postal survey of two random samples of parents of preschool children (aged 3 years). One sample (A) involved parents living in more affluent areas and the other (B) parents living in less affluent areas. RESULTS: In general, parents in both groups showed similar attitudes towards home safety. The only significant differences to emerge were over parental perceptions of the safety of the neighbourhood in which they lived and over the availability of money to keep their child safe (group B > group A, p

School injuries in an occupational health perspective: what do we learn from community based epidemiological studies?

Abstract: OBJECTIVE: The paper reviews community based epidemiological studies of injuries at school. It discusses their main findings in the light of a frame of analysis that emphasizes multidimensionality of causes, and in a Swedish context, where school injuries are legally regarded as occupational injuries. METHODS: A frame of analysis, inspired by research in the arena of occupational accident, was developed. It employs four key concepts to distinguish between injury and accident sequence, and between situational and structural factors as potential injury determinants. It also stresses the interaction between pupil, school, and community, in injury genesis. In parallel, a review of community based epidemiological studies of school injuries was undertaken after searching the relevant literature. The knowledge gathered from these studies is appraised in the light of the frame of analysis. RESULTS: In general, the studies point to potential risk groups, and to hazardous locations and activities. They address injury characteristics, but largely fail to define typical mechanisms. The literature sheds little light on whether there are particular characteristics of the community, school, or pupil(s)--other than gender and age--that influence injury event occurrence. CONCLUSIONS: Further research should devote attention to the sequences of events culminating in school injuries, and to the structural determinants of accident sequences. In this, surveillance systems might prove to be of great utility.

Cost outcome analysis in injury prevention and control: a primer on methods.

Abstract: Investments in public programs typically are constrained by a desire for fiscal responsibility. Decision makers are interested in knowing if an investment produces desired results less expensively than alternative approaches, or if an investment's benefits exceed its costs. They may want to determine whether a particular program is worthwhile to implement (a prospective approach), or whether a program that has already been implemented has been worth its cost (a retrospective approach). World wide, concerns about health care costs have pressed these issues to the forefront. Cost outcome analyses generally develop a measure of the cost per positive outcome from an intervention. By expressing outcomes in a common metric, such analyses often clarify murky resource allocation decisions. For example, is it more important to fix the swings, which will prevent five broken arms a year, or the seesaws, which will prevent four sprained ankles and two broken legs a year? Should we flatten the curve on High Street, which will prevent one pedestrian death every four years, or add a shoulder on Rose Street, which will prevent five hospitalizations a year? Besides helping to compare different interventions, cost outcome analyses help to identify the consequences and costs of a particular intervention. An injury prevention program may not only lead to the avoidance of injury and death and associated medical costs, but also a reduction in property damage, work loss, and pain and suffering. The costs of implementing the program include not only direct expenditures on salary, equipment and space, but also other uncosted resources such as volunteer time or public resources such as police time. Program costs and outcomes may sometimes be broken down into who actually bears the burden: health care providers, potential victims of injury, or taxpayers through additional government costs. To deal with such challenging questions, three types of cost outcome analyses (table 1) are available: Cost effectiveness analysis (CEA) expresses the outcome in a convenient and useful measure, for example, per life saved or per scald burn prevented. The findings are normally expressed as ratios, such as the cost per year of life saved or the cost per injury avoided. Cost utility analysis (CUA) extends cost effectiveness analysis by including different uncosted outcome measures, weighted by a common unit. The common unit is usually a quality adjusted life year, or QALY.1 QALYs (and variants, like the World Bank's disability adjusted life years), are scales that value a year in any given health state between death (with value 0) and perfect health (with value 1), based on a representative individual's preferences among different health states. QALYs reflect not only years of life saved but also the degree of functioning and health during those years. They do not reliably measure out-ofpocket cost savings due to care. Thus, it is generally desirable to subtract these savings (for example, reduced property loss and medical costs savings) from the cost term when computing a cost utility measure. Cost benefit analysis (CBA) places dollar values on all significant outcomes, including death, pain and suffering, and property loss, so that benefits are directly compared with costs in monetary terms. Reporting costs and outcomes in a common metric facilitates comparison over diverse programs, and allows the benefits to be clearly distinguished from the costs. (A CUA may be translated into a CBA by placing a dollar value on QALYs.'-5) We describe the steps in a cost outcome analysis, including the choice of perspective and the injury cost components relevant when valuing injury prevention. A benefit cost analysis of smoke detectors illustrates the methodology. We then discuss common errors in cost outcome analysis, and conclude with some recommendations.

Understanding suicide among indigenous adolescents: a review using the PRECEDE model.

Abstract: AIM: To use the available literature to identify the causes of suicide among indigenous adolescents. METHOD: The PRECEDE model provided a framework to organize the material and identify the areas where relatively little research had been reported. RESULTS: The epidemiological diagnosis showed that suicide was greater in indigenous than non-indigenous populations and particularly high among adolescent males. Environments of native persons are characterized by remoteness, poverty, cultural displacement, and family disintegration. The educational and organizational diagnosis identified predisposing factors reflecting the social environments previously identified, the enabling factors of televised suicides, and firearm and alcohol availability, in conjunction with an absence of positive expectations. Finally the administrative and policy diagnosis identified a piecemeal, short term perspective, often lacking cultural sensitivity. Although there was more literature from the United States than from Canada, Australia or New Zealand, the pictures emerging were consistent, with problems being identified across continents. Literature was more abundant in relation to the epidemiological, environmental, and educational/ organizational diagnoses than in relation to policy and administration. CONCLUSION: The increased suicide rates among indigenous adolescents were not a product of their native origins, but of the social milieu in which these people generally found themselves.

Pediatric injuries in an Arabian Gulf country.

Abstract: OBJECTIVE: To determine the common types of injuries among children (0-14 years) in Al-Ain, United Arab Emirates (UAE). DESIGN: A retrospective descriptive hospital based study. SETTING: Al-Ain Medical District, Al-Ain Teaching Hospital, UAE. SUBJECTS: All patients aged 0-14 years seen at Al-Ain Teaching Hospital for injuries during 1994. RESULTS: The number of children with an injury who attended the emergency room was 16,518 (69.9% boys; 30.1% girls). Injury rates were higher among non-UAE nationals. The most frequent reason for hospital admission was poisoning (41%). In the age group

Evaluation of a community based childhood injury prevention program.

Abstract: OBJECTIVES: This pilot study evaluates the effectiveness of a community based childhood injury prevention program on the reduction of home hazards. METHODS: High risk pregnant women, who were enrolled in a home visiting program that augments existing health and human services, received initial home safety assessments. Clients received education about injury prevention practices, in addition to receiving selected home safety supplies. Fourteen questions from the initial assessment tool were repeated upon discharge from the program. Matched analyses were conducted to evaluate differences from initial assessment to discharge. RESULTS: A significantly larger proportion of homes were assessed as safe at discharge, compared with the initial assessment, for the following hazards: children riding unbuckled in all auto travel, Massachusetts Poison Center sticker on the telephone, outlet plugs in all unused electrical outlets, safety latches on cabinets and drawers, and syrup of ipecac in the home. CONCLUSIONS: A community based childhood injury prevention program providing education and safety supplies to clients significantly reduced four home hazards for which safety supplies were provided. Education and promotion of the proper use of child restraint systems in automobiles significantly reduced a fifth hazard, children riding unbuckled in auto travel. This program appears to reduce the prevalence of home hazards and, therefore, to increase home safety.

You can't believe all that you're told: the issue of unvalidated questionnaires.

Abstract: This column arose out of discussions about a mailed questionnaire to parents on home hazards and safety practices (see Evans et al, p29). The issue being discussed was the confidence that could be placed on the results in the absence of independent validation of questions and answers. One view was that to require formal validation is unrealistic. It was felt that such validation is onerous, and, even when resources are available, it is often difficult to know how best to validate such questionnaires. Those sharing this view also note that, while it is right to be wary of non-validated surveys, if scientific journals always required these data, few papers would be accepted. Mailed questionnaires are used because they are manageable, cost effective tools, but some are of little real value. Is it unfair to say that the direction and structure of some questionnaires are influenced more by what is convenient than by what is good technique? The pressure to ask questions that can be easily coded and expressed as numbers in a good looking table is very real. As one researcher suggested in conversation, research reports that contain plenty of percentages look meaningful, lend themselves to numerical quotation, and appeal to funding agencies. It is because there is almost always a compromise between science and resources that there are no solid answers about this issue. Accordingly, this opinion was commissioned to generate debate on what is good practice when conducting this type of research.

Occupational injury deaths of 16 and 17 year olds in the US: trends and comparisons with older workers.

Abstract: OBJECTIVE: To examine patterns of occupational injury deaths of 16 and 17 year olds in the United States for the three year period 1990-2, examine trends since the 1980s, and compare fatality rates with those of older workers. METHODS: Occupational injury deaths were analyzed using the death certificate based National Traumatic Occupational Fatalities (NTOF) surveillance system. Fatality rates were calculated using estimates of full time equivalent (FTE) workers based on data from the Current Population Survey, a monthly household survey. RESULTS: There were 111 deaths of 16 and 17 year olds for the years 1990-2. The average yearly rate was 3.5 deaths/100,000 FTE. The leading causes of death were motor vehicle related, homicide, and machinery related. All causes occupational injury fatality rates for 16 and 17 year olds were lower than for adults for 1990-2. Rates for the leading causes of death (motor vehicle related, homicide, and machinery related) were comparable or slightly higher than the rates for young and middle aged adult workers. Although rates decreased dramatically from 1980 to 1983, the decreasing trend attenuated in later years. CONCLUSIONS: Comparisons of youth fatality rates to those of adult workers should address differences in patterns of employment, most importantly hours of work. Comparisons to narrow age groupings of adults is preferable to a single category of all workers 18 years and older. Increasing compliance with federal child labor regulations could help reduce work related deaths of youth. Other measures are needed, however, as there are many work hazards, including those associated with homicides, that are not addressed by United States federal child labor law regulations.

Does sharing the cost of a bicycle helmet help promote helmet use

Abstract: OBJECTIVE: To determine whether asking for a $5.00 donation for bicycle helmets, compared with distribution free of charge, would affect helmet use among children receiving helmets and an educational intervention from public health clinics. SETTING: Six public health clinic sites in King County, Washington. METHODS: Six participating clinic sites were randomly assigned to either free helmet distribution or to a $5.00 suggested donation for the helmets, stratified by whether a helmet law was in place. Three sites were assigned to each arm. Children who were between 6 and 12 years of age and who reported riding bicycles, but having no bicycle helmets, were eligible. Clinicians distributed helmets and delivered an educational intervention to 506 eligible children, or siblings of children seen at the clinic between March and July 1993. Parents were contacted after helmet distribution to ascertain helmet use. RESULTS: 82% of children whose parents were asked for a copayment and 77% of children who received free helmets were reported to wear their helmets every time they rode their bicycles (p=0.20). The adjusted odds ratio for the association between copayment compared with free helmets and helmet use was 1.66 (95% confidence interval 0.94 to 2.92). CONCLUSIONS: Helmet use was not significantly different among children whose parents were asked for a small copayment, compared with those who received helmets free. Use of copayments can increase helmet use by increasing the number of helmets given to low income children.