Copyright 2000 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2000
To describe current research in child and adolescent injury prevention by pediatric and public health investigators for comparison with national recommendations and agendas.
Abstracts submitted to the 1998 annual meetings of the Pediatric Academic Societies/Ambulatory Pediatrics Association and the American Public Health Association on injury or violence in children or adolescents.
All abstracts of projects that addressed primarily injury or violence prevention involving children or adolescents.
For 123 abstracts, 2 coauthors extracted and classified age of the population, type of injury, study design, sizes of the sample and denominator, and type of outcome.
Adolescents were the most frequent (49%) age group included. The investigations were concerned most with injuries caused by violence (33%), followed by motor vehicle trauma (14%) and burns (7%). Descriptive surveillance (38%), surveys (32%), and case series (13%) comprised the overwhelming majority of methods used. The studies primarily sought to identify risk factors for injury (32%), describe the victims (20%), or measure knowledge and/or practice (26%). Nine studies (7%) sought to measure the effect of interventions in some way, and only 2 focused primarily on methodology development.
Injury prevention research projects presented at the 1998 Pediatric Academic Societies and American Public Health Association meetings were proportionate to the frequencies of injury by age and by external cause in the United States. However, in comparison with recommendations for agendas of national injury prevention research, more research is needed to improve injury prevention methods and to evaluate interventions.
SINCE THE 1985 report of the National Academy of Sciences entitled Injury in America: A Continuing Public Health Problem,1 research in the United States directed toward prevention of injuries has increased substantially.2- 5 Recognition of injuries as the major cause of death for children and adolescents has made prevention in this vulnerable group a high national public health priority. The major professional societies for pediatric research and public health have established special sections devoted specifically to the research and prevention of child and adolescent injuries. An international society and a journal of child and adolescent injury prevention have been successfully established. Numerous national meetings focus on childhood injury prevention. These activities and organizations attest to the fact that pediatric and public health communities recognize injury in children and adolescents as a major threat to their health that is worthy of considerable investigative activity. Defining directions for research, type of injuries addressed, needed methodologies, and questions to be asked, and the answers to be produced from this emerging field are important to maximize productivity and to ensure the greatest benefit toward preventing serious injury in children.
Two leading academic and professional societies, the Pediatric Academic Societies (PAS), including the Ambulatory Pediatrics Association, and the American Public Health Association (APHA), are concerned with research that addresses injury and injury prevention in children and adolescents. Each of these societies has a special interest group or a section through which most injury epidemiology and prevention abstracts are reviewed. Although these societies do not receive all of the presentations of childhood injury prevention research, they do represent important forums for the work that is currently conducted by the pediatric and public health community in child and adolescent injury prevention.
As chairpersons for the special interest group or section and reviewers of the child and adolescent injury abstracts, we thought that it would be informative to injury prevention researchers to summarize the type and focus of research projects that are currently being submitted in these forums. Such a summary provides an overview of the types of injuries being addressed and the approaches used by injury control researchers. This cross-sectional summary also presents an opportunity to compare the research being conducted with the actual risks for injury experienced by children and with the recommendations for research by various advisory groups and leaders in the field. In short, this summary looks at what and how are we doing in research on child and adolescent injury prevention.
All abstracts submitted to the PAS and the APHA Injury Control and Emergency Health Services or Maternal and Child Health Sections in 1998 that included injury or violence prevention topics in the title were reviewed for inclusion in this summary. Abstracts that included the child or adolescent ages were included in this analysis. Abstracts that addressed only adults or that did not address injury or violence as a major emphasis of the project were excluded. For example, we did not include studies of multiple risk-taking behaviors in adolescents where injury was only one of a number of other variables. Submitted abstracts that met the above criteria were included in this summary whether or not they were accepted for the program of the respective meeting.
Each abstract was classified by the following features: age of the population of concern for injury, defining characteristic of the study population or group, type of injury by either nature of injury or external cause, study design, sample size, denominator, and type of outcome expected. Since some projects addressed multiple aspects of each variable or attribute, where appropriate, multiple categories were used for age, population of interest, and type of injuries. For study design, sample size, type of denominator, and the principal outcome, nonoverlapping classification was used.
For each abstract, a reviewer (P.C.S., M.D.O., or T.C.) completed a form that categorized all of the above features. The data were subsequently converted to an SAS statistical software file (SAS Institute Inc, Cary, NC) for analysis and calculation of frequencies. Both chairpersons (P.C.S. and M.D.O.) reviewed all study outcomes and ambiguous classifications and concurred on a final classification. The frequency of categories for each variable was tabulated to identify the relative emphases in injury prevention research by this group of investigators in pediatric and public health childhood injury prevention.
To prevent overrepresentation by duplicate abstracts from a single project, we screened for abstracts submitted to both organizations that addressed the same age, activity of the population related to injury, and injury of focus. One pair of abstracts was identified as a duplicate submission to both organizations, and the later of the paired submissions was eliminated.
A total of 124 injury prevention abstracts involving children and adolescents were submitted for presentation at the 2 annual meetings. Eliminating 1 abstract submitted to both organizations left a total of 123 projects, 60 submitted to the PAS and 63 to the APHA. They composed 1.6% of the almost 4000 APHA abstracts and 2.5% of the 2353 total PAS abstracts. However, within the PAS, most of the childhood injury prevention abstracts were submitted to the Ambulatory Pediatrics Association, where they composed 16.9% of the 365 abstracts.
The age ranges encompassed by the projects varied from a single narrow age (eg, infants younger than 1 year) to multiple ages in various combinations. Therefore, Table 1 presents the number and percentage of projects that focused only on a single, relatively narrow age range (eg, only adolescents) and projects that included given age ranges either singly or in combination (eg, all school age). One fifth (25) of the projects encompassed all ages including adults, and 11 other projects included all pediatric and adolescent ages without the inclusion of adults. Of the studies limited to a specific age range, the largest number (16 [13%]) focused on injuries in the adolescent age group, and adolescents were the most frequent age range (48.8%) included in this group of research projects. Thus, these projects addressed a broad distribution of age ranges in childhood and adolescence.
Approximately 30% of the studies broadly included all injuries without specifying the nature of injury, cause, or related activity (Table 2). For the remainder of the studies, the injuries of interest were categorized by various definitions of the injury or event. Some projects focused on injuries designated by the mechanism of injury (eg, a fall); others, by type of activities related to the event (eg, riding a bicycle); and still others, by the nature of the injury itself (eg, fracture). In some projects there were combinations of these designations, such as a study that focused on intracranial injury specifically from child abuse. Such multicategorical projects are counted in each category; therefore, the sum of the studies listed in Table 2 is more than the 123 abstracts. Violence-related injuries composed the largest group, followed by transportation-related injuries and then various forms of burn injury.
The study design or type of project in Table 3 is ordered roughly by the strength of evidence of efficacy used by the US Preventive Services Task Force.6 Less than 10% of the projects undertook an evaluation of an intervention, and only 3 (2.4%) did so as either a randomized or a nonrandomized controlled evaluation. The majority of the projects (105 [85.4%]) were surveys, case descriptions, or some other form of descriptive surveillance, such as record reviews from emergency departments. Denominator populations ranged in size from a single city or municipality for 14 projects, to a state or states in 19, to the entire US population for 16 of the projects. Denominator data were not applicable in 20 studies (16.2%) and not used or available in 32 (26%) of the projects.
What outcomes did the investigators expect to achieve? Table 4 lists the apparent intended outcomes from evaluation of interventions to descriptions of features for a group of injured victims. The majority of projects (87.8%) sought to describe some aspect of injuries or related factors (trends, costs, knowledge or behavior, practice, risk factors, and case descriptions). More than half of the apparent outcomes consisted of the identification of risk factors for injuries (39 projects [31.7%]) and case descriptions of injuries, victims, or how they were treated (24 projects [19.5%]). Only 9 (7.3%) of the projects sought to evaluate the effectiveness of an intervention to prevent injuries, and only 2 (1.6%) aimed to develop or improve methodology as the focus of the study.
This snapshot of current childhood injury prevention efforts reveals that the energy and attention of researchers in child and adolescent injury prevention tend to be broadly based and inclusive of wide age ranges and multiple types of injuries. It is reassuring that adolescence, which is the age of highest risk for injury,7- 9 is targeted for the most investigative activity as a single focused age group and is the group most frequently included in studies of broader ages. However, adolescents are still underrepresented as the focus of projects in view of the relative frequency of injury to adolescents compared with other age groups. That violence-related injuries are most frequently targeted for study by external cause is consistent with the rise in violence during the past decade and the associated increasing concerns of society.
A limitation of this review is that the sources for pediatric injury prevention research projects were restricted to the 2 professional organizations. Accordingly, these data cannot be truly representative of the universe of research activities of child and adolescent injury prevention at the time. The 2 organizations are necessarily biased toward public health and clinical and epidemiological methods by the disciplines of their membership. A complete picture of the pediatric injury prevention research effort would require a thorough review of the literature, cataloging of government and foundation research support, and the inclusion of proceedings from other relevant meetings, such as the Society for Adolescent Medicine, international and national injury conferences, and the American Academy of Pediatrics Section on Injury and Poisoning. Other approaches, such as biomechanical mechanisms of injury or environmental reengineering for child safety, would be found mostly through other sources. To summarize these other sources is beyond the scope of this report and our resources. Even with this limitation, as feedback to investigators interested in child and adolescent injuries, we believed that it would be useful for this group to be aware of the current picture of clinical and public health research in relation to identified needs and recommendations for injury prevention.
How does this cross section of reported research match with recommendations and agendas for the research needed to further prevent injuries in children and youth? The Institute of Medicine Committee on Injury in America of the National Academy of Sciences1 recommended ongoing surveillance, biomechanical research that clarifies the mechanisms of injury and human tissue responses to stress, and evaluation of interventions with controlled intervention studies. In 1999, another Institute of Medicine committee further emphasized the importance of rigorous analytical methods to conduct randomized control trials, cohort studies, and case-control studies.10 A workshop of experts and leaders in childhood injury prevention and in related research11 recommended a research agenda specifically for childhood injury prevention. This agenda included the development of improved methods, such as severity measures, proxy measures for serious injuries, or measures of exposure to risk; more effective measures of environmental hazards; and methods to improve understanding of the processes of injury, such as simulations and the study of "near-miss" events. The panel emphasized the need to better understand the interaction of child development and environment, mechanisms of how risk and protective factors affect outcome, effective behavioral change for injury prevention, and community change for safe environments. The American Academy of Pediatrics Committee on Injury and Poison Prevention published an agenda of research to "reduce the toll of injuries" in children that included causal process analysis, improved environmental controls, and behavior change strategies in addition to surveillance, epidemiology, and rigorous evaluation interventions mentioned above.5 These are but a few of the published agendas and recommendations for effective advancement of child and adolescent injury control through research.12- 16
Although surveillance is the cornerstone of injury prevention,4,10,12,17,18 the current efforts of the pediatric and public health community are overwhelmingly and, we suggest, disproportionately dedicated to descriptive surveillance and epidemiology. Understandably, investigators undertake and respond to directions of research according to the avenues that are available and the resources to support them. However, in comparison with the recommended research agendas mentioned above, this summary of reported activity suggests a paucity of research directed to improve methods to study and evaluate childhood injury prevention. Furthermore, given the relative large number of studies submitted in 1998, there are surprisingly few evaluations of interventions for injury control. While not abandoning surveillance and epidemiology of the injury experiences of children and adolescents, the field would be well served by investigators increasing their attention and energy to the development of improved methods and to evaluation of the measures and interventions undertaken to control injuries.
Accepted for publication October 6, 1999.
Presented at the Annual Meeting of the Injury Control Special Interest Group of the Ambulatory Pediatrics Association, New Orleans, La, May 4, 1998.
We greatly appreciate Jonathan Kotch, MD, MPH, chair, Section of Maternal and Child Health of the American Public Heath Association, for his help in providing abstracts from the Section of Maternal and Child Health and for supporting this review.
Reprints: Peter C. Scheidt, MD, MPH, Children's National Medical Center, 111 Michigan Ave NW, Washington, DC 20010 (e-mail: firstname.lastname@example.org).
Scheidt PC, Overpeck MD, Trifiletti LB, Cheng T. Child and Adolescent Injury Research in 1998A Summary of Abstracts Submitted to the Ambulatory Pediatrics Association and the American Public Health Association. Arch Pediatr Adolesc Med. 2000;154(5):442-445. doi:10.1001/archpedi.154.5.442