10.4 Injury control: the public health approach
Oxford Textbook of Public Health
Injury control: the public health approach
Corinne Peek-Asa, Bonnie Dean, and Jess F. Kraus
Public health history of injuries
Causal model of injury control
Injury incidence and trends
Emergency department visits
Impairment and disability
Lost productivity and cost of injuries
International injury comparisons
Injury control strategies
Examples of injury control strategies
Organization of injury control activities
In the last decades, decreases have been observed in almost all causes of injury death, although disparities among age, sex, and race/ethnic groups remain. In the last 75 years, the motor vehicle fatality rate per mile driven has decreased 90 per cent, yet despite this decrease motor vehicles remain the most common cause of injury death. After increasing 22 per cent from 1985 to 1993, the firearm-related mortality rate decreased 11 per cent from 1993 to 1995. These trends indicate that although great strides have been made in preventing injury deaths, changing risk patterns and disparity in exposures continue to challenge the field.
Observed decreases are due in large part to increasing activity in the systematic study of the causes of injury, and the broad application of this knowledge to prevention. This activity is known in sum as injury control, which is a multidisciplinary effort driven by public health to reduce the magnitude, severity, and consequences of injuries. Increased injury control activity was enabled by a slow change in how we think about risk and prevention. The very term ‘accident’, the most common reference to injurious events, evokes a feeling of chance and helplessness. However, injuries occur in predictable and usually preventable patterns, and our challenge is to identify and alter these patterns.
Injuries affect all human beings at some level. In fact, most of us sustain injuries from a very young age and continue to suffer from them throughout our lives. Perhaps this acceptance of injury as a part of everyday life delayed our efforts at prevention. However, the toll of injuries could not be ignored forever. Injuries take too many young lives suddenly and unexpectedly. Severe injuries can lead to lifelong disabilities and chronic pain which can restrict work, recreation, and leisure activities temporarily or permanently. Injuries can tear families, businesses, and communities apart, as was demonstrated in disastrous events such as Hurricane Mitch and the shooting at Columbine High School in Colorado. Individuals and societies can be left with enormous medical bills, extensive rehabilitation, major lifestyle adjustments, and post-traumatic stress and depression—losses that cannot easily, if ever, be recouped.
Although public awareness often focuses on events that cause many injuries at once, the toll of injuries climbs steadily every second of every day. The majority of injuries do not occur in mass events; for example, the annual number of deaths from motor vehicle crashes far exceeds that from airline crashes and natural disasters combined. Injuries disproportionately affect younger people more than the better-recognized threats to health such as cancer and circulatory disorders. The general public is unaware that injuries are a leading contributor to years of premature life lost, and missed days of work and school, and are one of the largest components of the medical care dollar spent per capita.
Despite successes in many areas of injury prevention, new risks and increasing population size constantly challenge the injury control community. Effective injury prevention strategies involve professionals from many fields working together. Modern injury control research combines ideas and skills from public health, biomechanics, engineering, behavioural sciences, law, law enforcement, medicine, and urban planning, among others.
This chapter will present a short public health history of injuries, examine the magnitude and distribution of injuries in the United States and the world, and outline approaches to injury control countermeasures. Injury control is a broad and complex field. This chapter will focus on injury prevention using public health models and how these models fit into the multidisciplinary field of injury control.
Public health history of injuries
Although injuries have been a leading cause of death throughout the history of humanity, they were not systematically studied until the twentieth century. The first formal safety organization, the National Safety Council, was chartered in 1913 with the primary goal of decreasing occupational fatality. The systematic study of injury incidence, risk factors, and successful prevention, however, was only beginning. One of the first epidemiological studies to document the effectiveness of an injury prevention strategy demonstrated that helmets decreased head injuries among motorcycle riders in the military (Cairns 1941; Cairns and Holbourn 1943). Cairns and his associates compared head injury incidence between helmeted and unhelmeted riders and were among the first to recognize the importance of studying defined populations using appropriate comparison groups.
John Gordon (1949) identified injury patterns by age, gender, and other demographic factors, noting that ‘accidents’ could be studied using epidemiological research methods similar to those in disease prevention. James Gibson (1961) defined energy and its many forms as the agent of injury. William Haddon Jr developed a framework for injury causation based on the infectious disease model (Haddon 1963; Haddon et al. 1964). He recognized that injuries occur when energy delivered to a living host from a vehicle or vector exceeds human tolerance, and classified the agent–host interaction into two categories of human–energy interaction: (a) when energy is delivered in excess of human tolerance, such as mechanical energy in motor vehicle crashes or in falls; (b) when energy metabolism is interfered with and physiological functions are impeded, such as in drowning or poisoning. Haddon used these basic ideas as a framework to develop a comprehensive matrix of host–energy interactions, discussed later in this chapter.
The study of injury prevention was greatly advanced by combining expertise from engineering with epidemiology. In 1941, Hugh DeHaven survived a crash of his trainer aircraft and noticed that his abdominal injuries were related to the shape and riveting location of his safety belt. He then studied ways in which vehicle engineering could reduce the severity of injuries during crashes (DeHaven 1968). His work bred new studies on human tolerance to energy forces during many types of impacts.
The ability to identify the incidence and causes of injuries in populations has been an important development in injury research. Agencies such as the National Safety Council, the National Highway Traffic Safety Administration, and the National Center for Injury Prevention and Control have been instrumental in establishing national surveillance systems for certain injury mechanisms. As the high incidence and cost of injuries have become better recognized, hospitals and communities have established local injury surveillance systems. The first documented computerized trauma surveillance system was introduced in 1969 at Cook County Hospital in Chicago (Pollock and McClain 1989). In 1980, Massachusetts and Ohio were the first states to introduce emergency room surveillance, but most states have not been able to maintain such comprehensive databases because of prohibitive costs. Some areas, such as San Diego County in California, are combining information from hospitals, the Emergency Medical Service Authority, and police to form a comprehensive surveillance programme. The National Highway Traffic Safety Administration has since implemented statewide programmes to incorporate crash information with medical outcomes. International collaborations have been underway over the last several years and are now providing information to compare risks and prevention efforts in many different environments.
Legislation has been an important element of injury prevention in the United States. Among the first legislative actions to reduce injury risk was a 1927 bill prohibiting interstate mailing of firearms through the United States Postal Service (Institute of Medicine 1999). Some highly successful legislation includes the 1953 Flammable Fabrics Act, the 1970 Poison Prevention Packaging Act, and the 1973 Emergency Medical Services Systems Act. While some legislative efforts have been long-lasting, others have had more contested histories. Speed limits, motorcycle helmet use laws, and gun control measures are examples of legislation which continue to be highly debated, with compromises difficult to achieve.
The level of injury control activity has grown remarkably over the last two decades, with increases in prevention programmes, research, and funding at federal and local levels. This activity has led to decreases in injuries from many causes. Further growth of injury control activities will be an important factor in sustaining and further reducing injuries.
Causal model of injury control
The traditional epidemiological model for infectious diseases provided the framework for early epidemiological studies of injury (Fig. 1). At the centre of the causal pathway for injuries is the agent–host interaction. The agent, which in the case of injuries is energy, is absorbed by the host to cause injury. Energy can take many forms, such as mechanical, electrical, chemical, radiation, and thermal. An example of an agent–host relationship is a motor vehicle crash, in which the energy exerted on the individual is mechanical. The reservoir is the place in the environment where the agent is found. The potential for energy transfer exists everywhere, but its ability to cause injury is limited. For instance, the potential energy in a bullet causes injury only when the bullet is in motion and hits a human.
Fig. 1 The causal pathway for injuries.
Vehicles and vectors are mechanisms which transport energy from the reservoir to the host. A vehicle is an inanimate object, such as a motor vehicle; a vector is animate, such as the dog that bites a child. For many injury causes, vehicles and vectors are both involved in energy transfer, such as when one individual (vector) stabs another with a knife (vehicle). The injury outcome is the trauma or injury sustained by the individual, and is influenced by host responses to the energy. Only energy transmitted beyond a host’s tolerance causes an injury, and therefore not all exposures to energy result in noticeable injury. A human has some resistance to energy which can be increased through exercise or protective devices, or reduced through changes in intrinsic factors such as existing medical conditions or age, or through extrinsic factors such as fatigue and alcohol.
This model presents an easily understood framework for the causal pathway of njuries. However, it can appear deceptively simple. The interactions of vehicles, vectors, and hosts within the physical, social, economic, and cultural environment, as well as the factors which influence tolerance of the host and how severe the consequences of injuries to the host become, are extremely complex and ever-changing.
Comprehensive and accurate data sources are the most important component in the cycle of surveillance, risk factor assessment, prevention, and evaluation. Data sources describing injuries are available on many levels, from broad national surveillance systems to local initiatives. In general, the detail in information provided decreases with the size of the surveillance system.
Increased efforts in injury surveillance are an exciting advancement in the field of injury control. A recent review by the Institute of Medicine (1999) identified 31 federally funded surveillance systems which address the incidence of injuries. The emergence of the Internet and improved computing power have made these databases easier to use and more accessible.
Specific data sources are described below.
Incidence data of fatal injuries in the United States are available from 1900 through the present in the Vital Statistics Records collected by the National Center for Health Statistics. These data are collected from death certificates and classified by E-codes. The United States Vital Statistics Records are a good source for counts of fatal injuries by broadly defined causes as well as by age, race, and gender, but detailed information about the event and the types of injuries sustained is not available. Injury mortality rates can be obtained through CDC Wonder, an internet site maintained by the Centers for Disease Control and Prevention (http://www.wonder.cdc.gov/). On-line queries of many databases can be conducted with instant results.
Most countries have some form of death certificate which is the source of official mortality counts. Deaths in developed countries have been accurately reported for many years. Mortality counts in less developed countries are less reliable and are often not available by cause. Although international comparisons of mortality statistics are routinely reported, the ability to compare international rates is limited by differing age distributions and exposures, as well as differences in reporting and coding criteria and data reliability.
In addition to overall injury mortality, several databases addressing specific causes of injury are maintained. A complete United States database on motor vehicle crash-related fatalities is found in the Fatal Accident Reporting System, developed in 1975 by the National Highway Traffic Safety Administration of the United States Department of Transportation. These standardized data on fatalities are collected from state and local police agency crash reports. Beginning in 1988, the General Estimates System was added to the Fatal Accident Reporting System database. The General Estimates System uses a nationally representative probability sample selected from all police-reported cases to estimate annual non-fatal crash injuries in the United States.
Work-related fatal injuries are collected by several sources. The National Traumatic Occupational Fatality System developed by the National Institute for Occupational Safety and Health was the first national system and uses United States death certificates for case identification. Work-related fatalities are determined by one item on the death certificate that indicates if the death occurred while the decedent was at work. The validity of this item to identify work-related fatalities varies by occupation, age, and gender. The specificity in correctly identifying all work-related fatalities may be low, leading to underestimates of work-related fatalities (Kraus et al. 1995). Since 1992, the Bureau of Labor Statistics has maintained the Census of Fatal Occupational Injuries. This census collects information from each state and may be the most accurate system currently in place because multiple data sources are used to identify and classify work-related injury. The National Safety Council also collects information on work-related fatalities. All three sources of work-related fatal injury data use different definitions of work, collect information from different populations of workers, and use different denominators for rate calculations.
National estimates of homicide deaths in the United States beginning in 1976 are maintained by the Federal Bureau of Investigations Uniform Crime Reporting System. Information on the victim, incident, and offender involved in the homicide are included. Data through 1996 are available to download from the Bureau of Justice Statistics web-page (http://www.ojp.usdoj.gov/bjs). Queries of databases describing perpetrators of crime are available on the same website through the Federal Justice Statistics Resource Center.
National estimates of injury morbidity are available from the National Health Interview Survey. Responses from sampled household interviews are weighted to estimate the incidence of all injuries in the United States, regardless of medical care received. Although this telephone interview provides valuable prevalence information, household telephone surveys often undersample severely injured individuals.
Information on hospital admissions, required for all accredited hospitals, is a source of national and community estimates of injury incidence. The National Hospital Discharge Survey conducted by the National Center for Health Statistics collects discharge data from each state to estimate national counts of injuries which require hospital admission. Discharge data are also usually available at the state and community level. These data represent only the most severe injuries. National estimates of emergency department visits and physician’s office visits for injuries are available through surveys conducted by the National Center for Health Statistics.
Population-based surveys are conducted for information about risk-taking behaviours and injury events. One example is the National Health Interview Survey, which measures many aspects of health status including a few variables addressing injuries. The National Crime Victimization Survey is conducted annually to determine incidence and outcomes from crime victimization. The Behavioral Risk Factor Survey determines changes in perception of risk and the prevalence of behavioural risk factors such as wearing helmets and using seatbelts. The National Nursing Home Survey provides information about injuries sustained by older persons residing in nursing homes.
Many surveillance systems provide detailed information about specific types of injuries, exposures, and outcomes. The National Electronic Injury Surveillance System conducted by the United States Consumer Product Safety Commission gathers information about product-related injuries requiring hospital admission or emergency department treatment from a national sample of hospitals. National surveillance systems exist for burns, homicides and assaults, medicare claims, and child abuse and neglect. Periodic regional surveillance is commonly conducted for firearm injuries and traumatic brain injuries.
Most countries now have mortality statistics which identify injury deaths, and these have been used for new efforts in comparative analyses (see the discusion of international injury comparisons below). As these collaborative efforts progress, disparities in coding and reporting practices will be improved. Many countries which have an integrated hospital system have comprehensive injury surveillance programmes. For example, the New Zealand Health Information Service of the Ministry of Health has collected detailed information about all individuals admitted to a hospital in the country since 1979 (Langley 1995). Australia conducts the National Injury Surveillance Programme, which compiles injury hospitalizations for the entire country, and also conducts the National Health Survey, which is comparable with the United States National Health Interview Survey (McClure 1995). Canada has morbidity statistics available from 1979. Specific efforts in injury surveillance began in 1989 by co-ordinating emergency-room-based reporting from 10 Canadian paediatric hospitals. This programme has since grown to include all age groups and more hospitals (Sherman 1995). Sweden established a national discharge register in 1964, and began the National Injury Prevention Programme in 1986. This programme does not yet represent all counties of Sweden (Berg et al. 1995). Collaborative efforts among countries are leading to increased comparability and reliability among existing surveillance strategies, and helping establish surveillance systems in countries which do not yet have them.
Many features of injuries are necessary for a complete understanding of their causes, and the coding of injuries has been a complicated issue. The cause and nature of the injury as well as the acute and chronic physiological damage are each important issues. Injury causes are usually classified according to the International Classification of Diseases External Cause of Injury Codes (E-codes), and have traditionally been categorized as either unintentional or intentional. Unintentional injuries include motor and non-motor vehicle occupants, pedestrians, drowning, poisoning, falls, and suffocation, among others. Intentional injuries have included homicide and suicide. This system, however, combines cause with intent. For example, suicide, an intentional injury, can occur through drowning or poisoning. Recent improvements in injury classification schemes have separated intent and cause, and a matrix which separates E-codes by cause and intent is available (Fingerhut and Warner 1997). This classification approach is included in the 10th revision of the International Classification of Diseases. In this system, homicide and suicide are no longer causes of injury but are an intent. Firearms are now a cause of injury, for which the intent is divided into homicide, suicide, and unintentional. This system is much more informative and provides categories which are mutually exclusive.
There are two standard methods by which anatomical injuries are coded. The International Classification of Diseases system provides a complete taxonomy of injuries, categorized by anatomical location and type of damage. This system is used by most hospitals for their discharge data systems. However, the International Classification of Diseases system does not provide any information about the severity of injury. The Abbreviated Injury Scale (Association for the Advancement of Automotive Medicine 1990) categorizes injuries by the anatomical location and nature of damage, but also includes an index for injury severity. Scores for each individual injury can be combined through a defined algorithm to produce the Injury Severity Score, which is a single estimate of overall body damage from a given injury event. The Injury Severity Score was developed as a predictor of mortality in automotive crashes, but has become the main system for classifying overall injury severity.
Classifying long-term impairment and disability is very complex. Impairment refers to physical damage, such as loss of range of motion or inability to bear weight on a joint. Disability refers to the inability to perform tasks. The overall disability for an injured individual depends on many factors in addition to the physiological damage, such as occupation, activity level, and health status. For example, someone who lifts heavy objects at work will have a greater disability from a weight-bearing impairment than someone who does not need to perform lifting tasks. Impairments and disabilities occur on such a large spectrum that it is difficult to classify them in one scale. The Activities of Daily Living scale measures performance of the most basic tasks, such as feeding, transferring, and toileting and is often used to measure very severe disabilities. One of the most widely used measures of disability is the Functional Independence Measure, which includes at a minimum 13 motor and 5 cognitive measures rated on a seven-level scale (Granger 1998). The Functional Capacity Index was developed as a companion to the Abbreviated Injury Scale, and provides an estimate of the expected reduced functional capacity for each injury in the scale (MacKenzie et al. 1996). One advantage of the Functional Capacity Index is that independent assessments post-injury are not necessary.
Existing systems of injury classification have improved with increased knowledge about the causal pathway of injuries. The area of injury classification has made much progress, yet also holds many opportunities for further development.
Injury incidence and trends
In 1997, unintentional injuries were the fifth leading cause of death in the United States, following heart disease, cancer, stroke, and pulmonary diseases. Suicide and homicide were the ninth and thirteenth leading causes respectively. However, unintentional injuries were the leading cause of death for those aged 1 to 44 years. Homicide was among the three leading causes of death for those aged 1 to 34 years, and suicide for the ages of 10 to 34 years. The three leading causes of death for those aged 15 to 24 years were unintentional injuries, homicide, and suicide (Hoyert et al. 1999).
Overall injury death rates have been decreasing since 1970, with the majority of the decline occurring among unintentional injuries. Figure 2 shows the change in the percentage of injury deaths by cause from 1987 to 1997. Although the rate of injury death has generally been decreasing, the proportionate causes of these deaths remained fairly stable. While unintentional injuries increased very slightly from 63.4 per cent to 63.9 per cent of all injury deaths, the proportion of motor-vehicle-related deaths decreased from 32.2 per cent to 29 per cent. This trend reflects a large decrease in the motor-vehicle-related fatality rate while other types of unintentional injuries, such as drowning and poisoning, have remained stable or increased slightly (Fingerhut and Warner 1997; Hoyert et al. 1999). The majority of the decrease in motor vehicle fatality rates has been for motor vehicle occupants, with less of an effect for other road users such as pedestrians. From 1987 to 1993, the motor vehicle crash rate decreased 19.4 per cent but remained relatively constant from 1993 to 1997. While motor vehicle deaths have decreased overall, these decreases have been less significant among children and the elderly (Brenner et al. 1999; Rivara 1999). Proper use and placement of child car seats and restraint of children who are too small for the shoulder harness and lap belt to fit properly are emerging focus areas. Further decreases in the motor vehicle death rates will require co-ordinated efforts at improving vehicle and road safety, public awareness campaigns, and new innovations in reducing alcohol and drug-involved crashes.
Fig. 2 Changes in the proportionate distribution of fatal injuries by type, United States, 1987 and 1997.
Homicides decreased from 14.1 per cent of injury deaths in 1987 to 13.3 per cent in 1997. Homicide deaths began decreasing in 1993, although the reasons for the decrease are not completely described. While increases in homicide rates since 1970 were attributed in part to increases in related crime, urban poverty, and inner city strife, as well as more accurate reporting of cause of death, recent decreases in homicide rates are attributed to the growing economy, community-based prevention programmes, increased and more effective law enforcement activities, and anticrime legislation (Sherman et al. 1997). The observed decreases are probably due in part to combinations of all of these as well as other undocumented factors. Suicide death rates remained at about 20.5 per cent of all injury deaths from 1987 to 1997. The most common and lethal method of suicide is the handgun. Increases in suicide death rates since 1950 are due almost exclusively to firearms, with the firearm death rate increasing 8.6 per cent from 1985 to 1995 (Fingerhut and Warner 1997). Increased access to firearms may lead to increases in suicide rates even if the number of attempts is stable, because attempts are more lethal. Non-complete suicide attempts are more commonly associated with poisoning and laceration (United States Department of Health and Human Services 1995).
Injury death rates have distinctive age patterns (Fig. 3). Non-motor-vehicle unintentional injuries are the leading cause of injury death from birth to age 5 and again after age 35, although the causes of these deaths differ. For those under 5 years the leading causes are suffocation, drowning, and fire/burns while the leading cause in the elderly is falls. Motor-vehicle-related deaths are the leading cause of injury death between the ages of 5 and 34 years and peak between the ages of 15 and 24 years. Those under 14 are more often killed as occupants in vehicles or as pedestrians, while the peak between the ages of 15 and 19 is largely due to the high fatal crash rate among younger drivers. Alcohol use is an important risk factor in motor vehicle crashes for all ages, but is a particular concern among drivers under 25. The elderly also have high pedestrian injury death rates.
Fig. 3 Injury mortality rates by age, United States, 1997.
Homicide is the second leading cause of injury death among infants and those aged 15 to 24 years. Although homicide has been decreasing in the last 5 years, these decreases have not occurred among those less than 18 years of age. Most homicides are committed with a firearm, occur during an argument, and are between acquaintances (United States Department of Health and Human Services 1995). Suicide rates are minimal until the age of 15 years, then increase steadily with increasing age. Trends indicate a reversal in the ranking of homicide and suicide deaths for persons over age 25, with suicide rates exceeding homicide rates.
Males have higher age-adjusted injury death rates than females for all causes of injury (Fig. 4). In 1997, the age-adjusted motor-vehicle-related death rate was approximately 53 per cent higher in males than females and showed similar gender variance in all age groups over age 5. The suicide rate was 76 per cent higher in males than in females, but diverged over age 65 when rates among males increased but for females remained constant. The homicide rate was 73 per cent higher for males than females. Homicide rates peak for males between the ages of 15 and 24 but for females peak between the ages of 25 and 34. While males are more likely to be killed by an acquaintance or stranger, females are overwhelmingly more likely to be killed by an intimate partner (Rennison 1999).
Fig. 4 Age-adjusted mortality rates for injuries by type and sex, United States, 1997.
Death rates for injuries also vary by racial group. In 1995, Native Americans had the highest unintentional injury death rate and the highest suicide rate (Fingerhut and Warner 1997). African-Americans had the highest homicide rate, which was 123 per cent higher than the rate for Hispanics, who had the second-highest rate. Asian-Americans had the lowest death rates for both intentional and unintentional injury.
Recent improvements in the classification of injuries separate intent from mechanism. Figure 5 shows the leading causes of injury death rates in 1997 by mechanism. Motor vehicle deaths were predominantly unintentional with a small number of suicides. Over half of all firearm deaths were suicides and another 40 per cent homicides. Less than 5 per cent of firearm injuries were unintentional, but these occurred primarily among the young. Approximately 28 per cent of poisoning deaths were suicides, and this proportion increased with age. Among the young, poisoning deaths were mostly unintentional. Over 14 per cent of poisoning deaths had an unknown intent. Falls, drowning, and fire/burn deaths were predominantly unintentional. The intent for suffocation was most often suicide, but when examined by age, homicide is the main cause of suffocation for those under the age of 1 year, unintentional suffocation was most frequent among those between 15 and 44, and suicidal suffocation increased with age. Cutting/piercing deaths were primarily homicides.
Fig. 5 Leading causes of injury mortality by intent, United States, 1997.
The injury pyramid is a common method for examining levels of injury severity including deaths, hospital admissions, emergency department visits, and office visits (Fig. 6) (McCaig and Stussman 1997; Woodwell 1997; Peters et al. 1998; Graves 1999). In 1996, the National Hospital Discharge Survey indicated that 2.55 million people were hospitalized in the United States for injuries (Graves 1999). This was a ratio of 17 hospital admissions for every death. For every death, 232 injured individuals (almost 35 million annually) were treated and released from hospital emergency departments and another 582 (almost 87.6 million annually) were treated in physicians’ offices (McCaig and Stussman 1997; Woodwell 1997). The sum of all levels of the injury pyramid yielded over 125 million injuries requiring some level of medical care in 1996. In comparison to 1992, the ratio of hospital admissions and emergency department visits for every death has decreased, but the ratio of physician office visits increased. This pattern suggests either a reduction in the severity of injuries or changes in treatment practices. Reductions in injury severity could reflect effectiveness of secondary prevention measures aimed at reducing the severity of injuries once a potential injury-producing exposure has occurred. Increases in the use of outpatient care could also contribute to decreases in hospitalizations.
Fig. 6 Injury pyramid of treatment levels and mortality, United States, 1996.
Injuries were the fourth leading cause of hospitalization in 1996, following heart disease, respiratory disease, and digestive system diseases (Graves 1999). For about 8 per cent of all individuals hospitalized, or approximately 2.55 million persons, the primary reason for hospitalization was an injury.
Table 1 shows the first-listed primary injury diagnosis and length of stay among hospital admissions from the National Hospital Discharge Survey (Graves 1999). The first-listed diagnosis is principally used for billing purposes and is usually but not necessarily the most severe injury sustained, and approximates the total number of individuals injured. However, first-listed diagnoses do not reflect multiple injuries and therefore underestimate the total number of individuals sustaining particular types of injuries each year, in particular the less severe injuries.
Table 1 Rate per 10 000 population of first-listed injury diagnosis on hospital discharge data and average length of hospital stay, United States, 1993–1994
The overall rate of hospital discharge for injuries in 1996 was 96.6 per 10 000 population with an average length of stay of 5.4 days. Fractures were the most common injury, with a rate of 38.9 per 10 000 population in 1993 to 1994, and required the longest average length of stay at 7.2 days. Poisoning, the second most frequent injury, occurred at a rate of 8.7 per 10 000 population and resulted in an average of 3.1 days in the hospital. Internal injuries, with a rate of 3.3, required the second longest hospital stay of 6.8 days. The rate of hospitalization for intracranial injuries was 5.9 and required an average of 6.5 days in the hospital.
Emergency department visits
Of the 94.9 million emergency department visits in the United States in 1997, 37 per cent were for injuries (Nourjah 1999). In addition to the fact that more injuries were presented to emergency departments than hospitals, the percentage of all emergency department visits attributable to injuries was far greater than the percentage of hospitalizations due to injuries. The rate of injury-related emergency room visits was 1311.9 per 10 000 persons per year.
Figure 7 shows the five leading injury-related complaints presented to emergency departments. About 6.4 million individuals were treated for falls, the leading cause of injury among emergency room visits. Over 4.8 million persons, or 13.7 per cent of all persons visiting the emergency department for injuries, were treated for injuries resulting from unintentionally being struck against or by objects or persons. More than 4.2 million people (12.2 per cent) were treated for motor vehicle crash injuries, almost 2.8 million (8.9 per cent) for unintentionally being cut by sharp or piercing objects, and an additional 2.16 million (6.1 per cent) for intentional-assault-related injuries.
Fig. 7 Leading causes of emergency department visits for injuries, United States, 1997.
The mechanism causing emergency-department-attended injuries varied by age (Fig. 8) (Burt and Fingerhut 1998). From 1992 to 1995, falls were the leading cause of emergency room visits for children under the age of 14 and for adults aged 65 and over, and were a common cause among all age groups. Motor vehicle crashes, the leading causes of death in persons between the ages of 15 and 34, were also the leading cause of emergency department visits for persons between the ages of 15 and 24. Being struck by or against sharp objects was not a common cause of injury death at any age but was among the top three mechanisms of injury treated in emergency departments for all age groups.
Fig. 8 Emergency department visits by mechanism of injury and age, United States, 1997.
The primary diagnosis for injury-related emergency visits was open wounds, which were listed as the primary diagnosis for over 5.7 million individuals and represented 26.1 per cent of all injury-related visits. Sprains, which constituted 20.3 per cent of all injury-related visits, were the second most commonly listed primary diagnosis, followed by contusions (15.9 per cent) and fractures (12.5 per cent). The distribution of injuries presented to emergency departments differed from the distribution of injuries requiring hospital admission because injuries treated in the emergency department were generally less serious.
Impairment and disability
The average annual prevalence of physical impairments due to injuries in the United States in 1985 to 1987 was almost 19 million, or 80 per 1000 persons (National Center for Health Statistics 1991). Individuals with moderately severe motor vehicle crash injuries had an average of 6.5 years of resulting impairment and 2.7 years of lost productivity. Over 26 per cent of all physical impairments were due to injuries, and impairments due to injuries compared with impairments from other causes disproportionately affect younger people (National Center for Health Statistics 1991).
According to the National Health Interview Survey for 1985 to 1987, over 70 per cent of impairments reported were deformities or orthopaedic impairments, which affect about 57 per 1000 population per year (National Center for Health Statistics 1991). Almost 43 per cent of orthopaedic impairments involve the back and 40 per cent involve the lower extremities. Amputations of extremities or parts of extremities, excluding tips of digits, represent about 7 per cent of impairment (5 per 1000 individuals), and paralysis represents 1.4 per cent (1 per 1000 individuals). Of the causes of impairments included in the National Health Interview Survey, motor vehicle crash injuries were the most common cause of impairment, followed by work-related injuries and injuries occurring in the home.
Disabilities due to brain and spinal cord injuries are generally the most severe and long-lasting. National estimates of the prevalence of persons permanently disabled due to brain and spinal cord injuries are not readily available. Because permanently disabled individuals often live in care-providing facilities, national domestic telephone surveys underestimate the prevalence of disability. Based on brain injury incidence and outcome estimates, the rate of new disabilities from brain injuries each year could be as high as 33 to 45 per 100 000 population (Kraus 1993).
Researchers are just beginning to measure the emotional trauma that impairment and disability have on loss of quality of life, self-esteem, and family structure. An increased public health focus on injury prevention will decrease the incidence of annual impairment and disability as well as improve quality of life for impaired individuals by increasing awareness about disabilities, modifying the environment to prevent injury, and increasing access and mobility for those with impairments.
Lost productivity and cost of injuries
The years of productive life lost from injuries exceeds all other life-threatening conditions (Rice et al. 1989). Although unintentional injuries are the fifth leading cause of death in the United States, they were the third leading cause of potential life lost in 1995 with 2.5 million years lost. When including suicide and homicide, the years of potential life lost to injuries exceeded 4.3 million (National Safety Council 1998).
Years of life lost due to premature mortality in the United States vary by gender and racial group. For males, African-Americans have the largest number of years of life lost before age 65, followed by Native Americans, Hispanics, non-Hispanic white men, and Asians. Among females, whose years of life lost are much less than males, non-Hispanic white females had greater losses than Hispanic females (Dosenclos and Hahn 1992).
Little is known about the degree of disability and quality of life for survivors of major trauma. Measuring years of functional life lost is complicated because so many factors must be considered, including functional status, pre- and postinjury employment, and life satisfaction. The years of functional life lost are greatest for survivors of brain and spinal injuries. In one study, only 55 per cent of patients who survived severe closed-head injury and who were employed full time before their injury could return to work (Stambrook et al. 1990). However, the greatest toll on lost work days is probably from minor injuries, which, although they result in relatively short absences, are highly prevalent.
Motor vehicle deaths are responsible for 37 per cent of life-years lost to injury, followed by firearms and falls, responsible for 22 per cent and 19 per cent of life years lost to injury, respectively. While the number of life-years lost to motor vehicle crashes has decreased in the last several years, the number due to firearm injuries is increasing, especially among individuals living in the inner cities of large metropolitan areas.
The economic costs of injuries can be measured in direct costs, those resulting from medical care, and indirect costs, such as those resulting from years of life lost, lost productivity, or property damage. In 1997, the cost of unintentional injuries, fatal and non-fatal, was calculated at $478.3 billion. This estimate included $238.4 billion in wage and productivity losses, $76.3 billion in medical expenses, $82.4 billion in administrative expenses, $21.3 billion in employer costs, and $59.9 billion associated with vehicle damage and fire loss (National Safety Council 1998).
Injuries are a significant contributor to health-care expenditures in the United States (Miller et al. 1994). In 1987, medical care spending for injuries was $69 billion (in 1993 dollars), accounting for approximately 12 per cent of total expenditures in health-care excluding insurance claims, dental care, and nursing home care (Miller et al. 1994). Cardiovascular disease was the only other category with a higher percentage of medical spending, capturing 14 per cent of the total costs. Hospitalizations from motor vehicle crash injuries were the most costly of all injuries, averaging $43 409 per injured person. Falls, burns, firearm injuries, and drowning each averaged over $30 000 per hospitalization (Rice 1989). The cost of medical care increases with increasing injury severity. For motor-vehicle-related injuries in 1990, the cost of medical care averaged $18 585 for a moderate injury, $57 030 for a serious injury, and $249 753 for a critical injury (National Highway Traffic Safety Administration 1992). Indirect costs also increase with injury severity because extended treatment needs are greater, there is more loss of productivity, and disabilities are more common.
Although fatalities often do not result in high medical costs, indirect costs can be extensive. Deaths which affect children have the highest costs because of lost productivity and the emotional trauma for the family. Indirect costs for injury events also include non-medical repair expenses such as property damage. In 1990, the total cost of motor vehicle crashes was $137.5 billion, which accounted for approximately 2.5 per cent of the United States gross domestic product (Blincoe and Faigin 1993). Of this cost, 75 per cent was associated with injury or fatal crashes. The overall cost of a motor vehicle crash per person was $6145 for minor injuries, $84 189 for serious injuries, $589 055 for critical injuries, and $702 281 for fatalities (Blincoe and Faigin 1993). While the predominant component of these costs for critical injuries was lifelong medical care, 80 per cent of the loss for fatalities was for productivity losses in the home and workplace.
International injury comparisons
International comparisons are only beginning to reveal the worldwide toll of injuries on premature death and lost productivity. Regardless of the country and its wealth or political practices, injuries are a leading cause of mortality and consistently affect the young disproportionately when compared with other causes of death. The extent to which injury incidence and causes differ across the world, however, indicate that the types of risk factors present vary substantially.
The World Health Organization’s Global Burden of Disease Study examined mortality in eight geographic regions defined by the World Bank (Murray and Lopez 1997a). It was found that one in 10 deaths worldwide were from an injury and that, overall, injuries are estimated to have caused over 5 million deaths in 1990. Among 107 causes of death, five injury causes were found in the top-ranking 25. The highest-ranking injuries were those due to road-traffic crashes, which ranked ninth as a cause for worldwide death and caused almost 1 million deaths worldwide in 1990. Self-inflicted injuries ranked twelfth, violence seventeenth, drowning twentieth, and war injuries twenty-first.
Developing countries had more than five times the injury deaths as developed countries, and more than 5.5 times the number of intentional injury deaths. The largest number of unintentional injury deaths were in India, China, and sub-Saharan Africa, and the largest number of intentional injury deaths were in sub-Saharan Africa, China, and the Middle Eastern crescent. Injury deaths showed regional variation by age and gender, and the Global Burden of Disease Study estimated that in China suicide was the cause of one in four deaths among women aged 15 to 44. These numbers do not control for the total population.
The Global Burden of Disease Study also estimated the burden of premature mortality and disability in the worldwide population (Murray nd Lopez 1997b). Disability was measured as disability-adjusted life-years (DALYs), which were calculated as the sum of life-years lost due to premature mortality and years lived with disability, adjusted for severity. Seven injury causes were among the 30 highest-ranked DALYs. These included road-traffic crashes (9), falls (13), war injuries (16), self-inflicted injuries (17), violence (19), drowning (21), and burns (27). Injuries caused 14.5 per cent of all DALYs in developed countries and 15.2 per cent in developing countries, which indicates that approximately 15 per cent of all premature death and years of disability are due to injuries. The highest DALY toll from injuries was found in the formerly socialist economies of Europe (18.7 per cent), China (17.6 per cent), and Latin America and the Carribean (16.4 per cent); the lowest were in established market economies (11.9 per cent) and the Middle Eastern crescent (13.0 per cent). Unintentional injuries contributed the highest proportion of DALYs in India (13.0 per cent), China (12.9 per cent), and the former socialist economies of Europe (12.9 per cent), and the lowest in the Middle Eastern crescent (6.8 per cent) and established market economies (8.7 per cent). Intentional injuries had the greatest contribution in established market economies (6.2 per cent), sub-Saharan Africa (6.0 per cent), and the former socialist economies of Europe (5.8 per cent), and the lowest in India (1.5 per cent).
Projections from the Global Burden of Disease Study indicate that injuries will increase in contribution as both a cause of death and a contributor to disease (Murray and Lopez 1997c). Deaths from injuries are expected to increase from 5.1 million in 1990 to 8.4 million in 2020. Road-traffic crashes are expected to increase from the ninth to the sixth leading cause by the year 2020, and other projected increases include self-inflicted injuries (twelfth to tenth), violence (sixteenth to fourteenth), and war injuries (twentieth to fifteenth). The contribution of injuries to DALYs is expected to increase to 21.1 per cent among developing countries and to 13.0 per cent for developed countries. The greatest increases are expected in sub-Saharan Africa, the Middle Eastern crescent, and India, with decreases expected in established market economies.
In 1994, the United States National Center for Health Statistics convened the first International Collaborative Effort on Injury Statistics, with the goals of comparing injury mortality rates and increasing comparability of reporting (Fingerhut et al. 1998). Eleven countries participated in this effort, all of higher development and socio-economic status. The International Collaborative Effort was able to apply the updated injury matrix to distinguish between cause and intent.
The highest injury death rates among the 11 countries were in France and Denmark, with 74.7 and 69.9 deaths per 100 000 population respectively (Fig. 9). England and Wales, Israel, and The Netherlands had the lowest injury death rates. Unintentional injury death rates were highest in France (48.2 per 100 000), Denmark (46.3), and Norway (41.4), and lowest in England and Wales (19.2). Suicide death rates were highest in France (20.8) and Denmark (18.4), and lowest in England and Wales (7.0) and Israel (7.0). Suicide by firearms was highest in the United States (7.0), which was the only country in which firearms were the leading cause of suicide. Poisoning was the leading cause of suicide in Australia, Denmark, England and Wales, New Zealand, and Scotland. Suffocation was the leading cause of suicide in Canada, France, Israel, The Netherlands, and Norway. Homicide rates were almost four time higher in the United States than any other country at 8.6 deaths per 100 000. Homicide rates in other countries ranged from 1.1 in Norway and France to 2.3 in Scotland.
Fig. 9 International comparisons of injury mortality rates.
Motor vehicle traffic deaths were the leading or second leading cause of injury death in all 11 countries, with the highest rate in New Zealand (21.3 per 100 000) and the lowest in England and Wales (6.2 per 100 000). Among those aged 15 to 24 years, the motor vehicle death rate in New Zealand was 49 deaths per 100 000.
Firearm deaths were the second leading cause of injury death in the United States with a rate of 13.7 per 100 000, but were not in the top three causes of death for any other country. Firearm injuries had the lowest mortality rate for all known causes in England and Wales, Denmark, The Netherlands, New Zealand, Norway, and Scotland. However, among those aged 15 to 24 years, firearms were the second leading cause of death in Norway, Israel, and France. These rates were far lower than found in the United States.
Poisoning death rates were highest in Denmark (13.4 per 100 000) and lowest in Israel (0.7 per 100 000). Deaths from falls were much more frequent in Denmark (25.7 per 100 000) than in any other country. Among those aged 65 and over, though, falls were the leading cause of injury death in all countries except Australia, France, and Israel. Deaths from suffocation were higher in France (14.1 per 100 000) than in any other country. Death rates from drowning ranged from 1.1 per 100 000 in England and Wales to 4.7 per 100 000 in Norway. Both France (18.6 per 100 000) and Norway (16.4 per 100 000) had a high number of injury deaths with unspecified causes, which complicates comparisons across countries.
Injury control strategies
The main objectives of injury research are to prevent the occurrence of injuries and to reduce their level of severity. Limiting prevention strategies to any single aspect of the many causes of injuries is an ineffective and narrow approach; successful strategies will incorporate many countermeasures and involve many different professionals (Association for the Advancement of Automotive Medicine 1993). Rather than ‘accident prevention’, the goal of injury control is better conceptualized by focusing on a general downshifting of severity over the entire spectrum of injuries.
Unlike many chronic diseases, the agent and time of injury onset is almost always known and can be measured, and the mechanism of energy transfer from reservoir to host can be described. With several exceptions, injuries usually occur immediately after exposure and rarely have the long incubation or latent periods of many infectious and chronic conditions. Within the framework of the public health model (Fig. 1), the primary focus of injury control is to identify energy forces which cause injury, define mechanisms of human exposure, and to identify precisely where interventions can interrupt the causal pathway.
The Haddon Matrix, a model of the agent–host relationship in injury causation, was the foundation for the study of motor vehicle crashes and countermeasures for highway safety, and continues to be an applicable theoretical framework for injury prevention (Haddon 1963, 1972). The Haddon Matrix divides the timing of the injury event into three phases: pre-event, event, and postevent (Table 2). These phases correspond to the three levels of prevention defined by public health.
Table 2 The Haddon Matrix with illustrations from motor vehicle safety
Primary prevention, which occurs during the pre-event phase, prevents the injury event by eliminating the mechanisms of energy transfer or exposure. Traffic safety laws or vehicle modifications which prevent automobile crashes, fences around swimming pools which prevent submersion, trigger locks on guns, and safety caps on poisonous substances are all examples of primary prevention which reduce or eliminate the chance of exposure.
The goal of secondary prevention, which occurs during the injury phase, is to eliminate or reduce injury severity once an energy transfer has occurred. Motorcycle helmets, seatbelts, life vests, and bulletproof vests are examples of secondary prevention. While they do not prevent the event which causes injury, they do reduce the energy absorbed by the host. It is important to note that some of the most effective secondary prevention strategies do not eliminate all injuries. For example, the motorcycle helmet is very effective in reducing head trauma in motorcycle crashes, but is not effective in preventing trauma to other body regions (Kraus et al. 1994).
Tertiary prevention, which occurs in the postinjury phase, aims to reduce the consequences of the injury once an injury-producing energy transfer has occurred. Emergency and trauma care, as well as rehabilitation efforts, are examples of tertiary prevention. Some of the most important advances in injury control have been improvements in the early response and treatment of serious injury.
Each of these phases is influenced by three factors: the human (host), vehicles or vectors, and the environment, which can be separated into physical and social, with economic and cultural aspects. Table 2 shows the Haddon Matrix with examples of each phase.
Specific injury-prevention strategies can be divided into two very broad groups based on need for host actions. Passive intervention requires no input or action by the host and is usually accomplished by modifying the agent, vehicle, vector, or environment. Modifications in car design to improve brakes and increase the energy absorbed by vehicle components are two examples. Active intervention requires that the host take some type of action for the intervention to work. Seatbelts and helmets are examples of active intervention. Just as effective injury-control strategies must address multiple facets of injury occurrence, they should also incorporate active and passive intervention strategies to be fully effective. Passive intervention strategies are usually considered more effective, especially when compared with active interventions which require frequent or time-consuming action (Waller 1985). Air bags, for example, require no driver action, whereas seatbelts can only be effective when fastened by the occupant. However, the most effective crash protection occurs when both are available.
One often neglected aspect of injury control is evaluation. Because effective prevention strategies are broad, multidisciplinary, and sometimes expensive, it is important to identify the specific features or groups of features which are most effective in injury control. Success depends on many factors, including the characteristics of the community in which the programme is implemented. Because communities vary greatly, evaluations must be conducted to determine if the effects of a programme in one community will translate to a different one. Evaluations must combine surveillance of injury events with a complete understanding of the injury control programme.
Examples of injury control strategies
Motor vehicle occupant injuries
The reduction in motor vehicle fatalities and injuries has been a major success in injury prevention. The rate of motor vehicle fatalities per vehicle mile travelled has decreased by 72 per cent since 1950 (National Safety Council 1998). This decrease has occurred despite an increasing number of motorists and vehicle miles travelled. The main reason for such success has been a comprehensive approach to intervention, which has included legislation and enforcement, educational programmes, engineering, and environmental changes.
Reducing alcohol use while driving has been an important approach to primary prevention. The positive relationship between alcohol use and increased crash involvement has been well established (National Committee for Injury Prevention and Control 1989). The number of alcohol-involved fatalities decreased by 24 per cent between 1982 and 1993, but levels have since remained stable. Legislative intervention and public awareness have been the main strategies employed to reduce drinking and driving. Legislation has included increases in the minimum drinking age, lower maximum breath-alcohol concentration, zero tolerance for underage drivers, and increasing enforcement and punishment of existing laws (Graham 1993). Public awareness campaigns have included designated or alternative driver programmes and health campaigns to reduce per capita alcohol consumption. Further reductions in drinking and driving fatalities will be caused by focusing efforts on high-risk drivers, such as repeat offenders and adolescent drivers.
The 1966 Highway Safety Act was instrumental in decreasing highway fatalities. Highway design incorporated smoothed curves, a divided traffic stream, controlled entrances and exits, and energy-absorbing guard rails, all of which contribute to lower crashes per vehicle mile than other roadways (Graham 1993). Crashes with heavy trucks and high vehicle speeds, however, are still areas in need of prevention activities.
Vehicle modification has been an important contributor to both primary and secondary prevention. Seatbelt use reduces motor vehicle fatalities and serious injuries by between 40 and 55 per cent (Evans 1985; Hedlund 1985; Mackay 1985). The lap portion of the lap–shoulder belt reduces movement and prevents ejection, and the shoulder portion reduces high-energy contact with the car interior (Chorba 1991). The effectiveness of seatbelts rests with compliance to mandated laws requiring their use. The National Highway Traffic Safety Administration estimated that in 1988, 4573 lives were saved through the use of lap–shoulder belts, and that an additional 15 959 lives could have been saved if everyone had worn safety belts (Partyka and Womble 1989). The cost of introducing and enforcing a mandatory seatbelt-use law is estimated at $69 per life-year saved, which is considerably smaller than the costs of treating injured drivers (Tengs et al. 1995).
The introduction of air bags further reduced fatality rates. Air bag effectiveness in frontal collisions has been demonstrated through crash testing and observational studies (National Committee for Injury Prevention and Control 1989). Assessing the independent effect of air bags is complicated because they are so frequently used with seatbelts, but air bags are estimated to reduce driver fatalities by 19 per cent and driver fatalities in frontal crashes by 28 per cent (Zador and Ciccone 1993). The NHTSA estimated in 1984 that 100 per cent use of full front-seat airbags alone would reduce deaths in the United States by 6190, serious injuries by 100 000 and minor injuries by 256 000 (National Highway Traffic Safety Administration 1984). The average dollar savings from these injury reductions is 8.7 billion annually. At an estimated cost of 4 billion dollars to install front-seat air bags in all cars, the surplus savings is 4.7 billion dollars (Rice et al. 1989).
The introduction of air bags, however, changed the vehicle environment and led to new risk factors for injury. The first air bags deployed at very high speeds and led to fatalities among children placed in rear-facing car seats in the front passenger position. There were also reported injuries to the elderly and drivers of small stature. Car seats used correctly and placed in the proper seating position decrease fatality and serious injury by approximately 71 per cent and 67 per cent respectively (Kahane 1986). The public health community mobilized quickly to introduce rear-seat placement campaigns, and engineers quickly began designing ‘smart’ air bags that can moderate deployment based on rider size and weight. Air bags are an example of the challenge of introducing a new prevention measure into an already complex environment.
Firearms are the second leading cause of death following motor vehicle crashes and are the leading cause of homicide and suicide deaths. Two issues must be addressed in firearm injury reduction: the lethality of firearms and their availability. The number of injuries resulting in death from firearms exceeds that of all other weapons, and firearms have two to five times the lethality of knives, the second most lethal weapon (National Research Council 1985). The availability of firearms may increase the rate of homicides, suicides, and unintentional firearm injuries, although this issue has been highly debated.
Successful firearm-related injury prevention requires multifaceted strategies which focus on high-risk individuals, the environment, and the firearm. Because of the lethality of firearms, the focus has remained on primary prevention. Firearm-related injury prevention has addressed the design, marketing, and legislation of firearms. Such a product-oriented focus has been used previously to reduce injuries associated with childhood poisonings and motor vehicles (Freed et al. 1998).
Primary prevention for firearm injury can be categorized into three areas (National Research Council 1985; Powell et al. 1996; Freed et al. 1998):
reduce the number and availability of guns in the environment
modify the use of the firearm or its storage
change the allocation through restrictions and licensing.
For each of these areas, prevention through legislative initiatives, technological changes, and training and education are feasible approaches.
Firearm availability can be reduced by limiting the type of weapons sold, the number of weapons sold to each person, or registration regulations. Voluntary approaches to reducing the number of firearms include gun buy-back programmes, educational campaigns, and increased prices and taxes. Because suicides are so often committed with a firearm, decreasing firearm availability is also a potential prevention strategy for suicide (O’Carroll et al. 1991).
The second strategy for reducing firearm injuries is to modify the use of the firearm and its storage. Many states have restrictions on carrying concealed weapons, but these have recently been challenged. Evaluations of mandatory sentencing for carrying without a proper permit and for use of a gun to commit a felony have demonstrated reductions in firearm-related homicides (National Research Council 1985).
Proper storage of firearms is necessary to reduce unauthorized or unintentional use, especially by children. One evaluation of a safe storage law found a 23 per cent decrease in unintentional shooting deaths among children younger than 15 years old (Cummings et al. 1997). Unauthorized use by children can be prevented by safety devices such as trigger locks on the firearm or lock boxes, and by increasing the strength needed to pull the trigger. New technologies which prohibit the firearm from discharging by unauthorized users are under development. Other modifications to reduce unintentional discharges include loaded chamber indicators, manual safeties, and magazine safeties.
A third approach to limiting firearm injuries involves changing the allocation of firearms through restrictions and licensing. Legislation to restrict use of firearms has been implemented since 1927 and has included licensing requirements, waiting periods, restrictions on sales to high-risk purchasers and the manufacturing of certain weapons, and disruption of illegal markets. Although there is debate about the effects of reducing availability of guns to the general public, there is general agreement that firearms should be kept away from potential criminals. However, methods of identifying specific individuals for regulation have not reached wide consensus. Although evaluations of legislation to deny purchase of a handgun to individuals with prior felony convictions indicate that the denial may reduce future criminal activity, more evaluation on this topic is needed (Wright et al. 1999).
Secondary prevention to reduce the lethality of firearms aims to reduce the severity of injury once a firearm has been discharged. Semi-automatic assault weapons, which allow rapid firing of many rounds of bullets, have been banned in the United States since 1989 (Powell et al. 1996). However, there are no generally accepted estimates of the number of assault weapons in the United States, their lethality compared with other weapons, or their use in crime (National Safety Council 1993; Powell et al. 1996). Debate over the precise definition of an assault weapon and the ability of gun-makers to make minor changes in banned guns to allow continuation of their sale have been complications.
Ammunition modifications are an important secondary prevention strategy. Some bullets are designed with full metal jackets and penetrate and exit the victim with reduced tissue destruction, especially when compared with bullets with deformable tips which balloon out on impact causing a wider path of tissue damage (Powell et al. 1996). Reductions in magazine size lower the number of bullets released from the firearm before reloading. Because ammunition is used rapidly when compared with the long functional life of the firearm itself, changes to ammunition may have a more immediate impact on reducing injuries (Freed et al. 1998).
Although the arena of gun control is ripe for debate, the high incidence of firearm injuries dictates this as a high priority for public health. The reduction of firearm injuries will require a co-ordinated effort with participation from many agencies and professionals, and its success will depend on continuing information from improving surveillance systems.
Great advances in the treatment of injured individuals have been made in the last 20 years, and have been a major contribution to decreases in death and disability. The trauma care system is a spectrum of services that includes prehospital care, acute hospital care, and rehabilitation. Advances in prehospital care include the emergency 911 system, poison control centres, and greatly enhanced emergency medical service field care and transportation with the advent of the Emergency Medical Services System Act in 1973 (Institute of Medicine 1997). The acute care phase was greatly enhanced with the development of trauma centres, which at the highest level have specialists and equipment immediately available to treat the most severe injuries (American College of Surgeons 1993). Networks of trauma systems work together to triage injured individuals to the appropriate level of care. Clinical research describing the physiological response to injury led to improved methods to stabilize and treat injuries and the importance of treatment in the first few hours after injury. The goal of rehabilitation is to minimize disability through functional improvements in the patient, which is related to efforts to improve the physical and social environments for the disabled (Institute of Medicine 1997). These three levels interact to provide co-ordinated and comprehensive care to victims of trauma. Future challenges include bolstering trauma care to rural areas and additional support of outcomes research to improve overall care.
Organization of injury control activities
In 1985, the National Research Council released the landmark report Injury in America, which identified priority areas for injury control activities and found that funding levels were not commensurate with the burden of the injury problem. Since that report several other national efforts to prioritize injury control activities have been conducted.
In 1990, The Healthy People 2000 report identified goals for national health improvement, including injury reduction. By 1996, overall injury death and hospitalization rates had reached objectives for 2000, but specific high-risk populations remained above the goals (United States Department of Health and Human Services 1995). Rates of homicide among young males, particularly among minority youths, suicide attempts among adolescents, weapons-related violent deaths, and non-fatal assault injuries were found to be far from year 2000 goals for reducing violence. The elderly and children, especially minority youth, were found to be far from goals stated for unintentional injuries. Child restraint and bicycle helmet usage, smoke detectors, and alcohol use were stated as priority areas. Currently, the United States Department of Health and Human Services (1998) is drafting health objectives for 2010. The number of injury-related goals has increased, and the new goals are focused on specific high-risk populations.
The Institute of Medicine has released several important reports addressing injuries and trauma care, the most recent of which was the report Reducing the Burden of Injury, produced in 1999. This report identified national priorities for surveillance, training and research, firearm prevention, trauma care systems, state infrastructure, and federal response. Recommendations from this report include expanding emergency department surveillance, the establishment of a national fatal intentional-injury surveillance system, expansion of training activities for both research and practice, and increased co-ordination and support for agencies leading the field of injury control.
Although these national priorities are vital to the growth and continued success of the field, the specific goals for prevention activities must occur locally. Each state, county, and community has different high-risk populations, injury risks, and attitudes towards prevention. These factors are a crucial element in the design of effective injury-prevention programmes. With increased recognition of the burden of injury as well as advances in our knowledge of the causes and consequences of injuries, the injury control field will continue to grow in an organized and informed manner.
Association for the Advancement of Automotive Medicine (1990). Abbreviated Injury Scale: revision. Association for the Advancement of Automotive Medicine, Des Plaines, IL.
Association for the Advancement of Automotive Medicine (1993). Injury control in the 1990s: a national plan for action. Report to the Second World Conference on Injury Control. Association for the Advancement of Automotive Medicine, Des Plaines, IL.
American College of Surgeons (1993). Resources for optimal care of the injured patient. American College of Surgeons, Chicago, IL.
Berg, L., Aberg, A., Schelp, L., and Svanstrom, L. (1995). Data needs for evaluation of injury prevention programs—experiences from Sweden. Proceedings of the International Collaborative Effort on Injury Statistics, Vol I, Chapter 19, pp. 1–9. Centers for Disease Control and Prevention, Department of Health and Human Services, Hyattsville, MD.
Blincoe, L.J. and Faigin, B.M. (1993). Economic impact of motor-vehicle crashes—United States, 1990. Morbidity and Mortality Weekly Report, 42, 443–8.
Brenner, R.A., Overpeck, M.D., Trumble, A.C., DerSimonian, R., and Berendes, H. (1999). Deaths attributable to injuries in infants, United States, 1983–1991. Pediatrics, 103, 968–74.
Burt, C.W. and Fingerhut, L.A. (1998). Injury visits to hospital emergency departments: United States, 1992–1995. Vital and Health Statistics, 13, 1–76.
Cairns, H. (1941). Head injuries in motor-cyclists: the importance of the crash helmet. British Medical Journal, ii, 465–71.
Cairns, H. and Holbourn, H. (1943). Head injuries in motor-cyclists: with special reference to crash helmets. British Medical Journal, i, 591–8.
Chorba, T.L. (1991). Assessing technologies for preventing injuries in motor vehicle crashes. International Journal of Technology Assessment in Health Care, 7, 296–314.
Cummings, P., Grossman, C.D., Rivara, F.P., and Koepsell, T.D. (1997). State gun safe storage laws and child mortality due to firearms. Journal of the American Medical Association, 278, 1084–6.
DeHaven, H. (1968). Beginnings of crash injury research. In Accident pathology (ed. K. Brinkham), pp. 1–15. FH 11–6595, Department of Transportation, Washington, DC.
Dosenclos, J.C. and Hahn, R.C. (1992). Years of potential life lost before the age 65, by race, Hispanic origin, and sex—United States, 1986–1988. Morbidity and Mortality Weekly Report, 41, 13–23.
Evans, L. (1985). The effectiveness of safety belts in preventing fatalities. General Motors Research, Warren, MI.
Fingerhut, L.A. and Warner, M. (1997). Injury chartbook: health, United States, 1996–97. National Center for Health Statistics, Hyattsville, MD.
Fingerhut, L.A., Cox, C.S., Warner, M., et al. (1998). International comparative analysis of injury mortality: findings from the International Collaborative Effort on Injury Statistics. Advance Data, 303, 1–18.
Freed, L.H., Vernick, J.S., and Hargarten, S.W. (1998). Prevention of firearm-related injuries and deaths among youth: a product-oriented approach. Pediatric Clinics of North America, 45, 427–38.
Gibson, J. (1961). The contribution of experimental psychology to the formulation of the problem of safety: a brief for basic science. In Behavioral approaches to accident research (ed. H.H. Jacobs), p. 77. Association for the Aid of Crippled Children, New York.
Gordon, F. (1949). The epidemiology of accidents. American Journal of Public Health, 39, 504–15.
Graham, J.D. (1993). Injuries from traffic crashes: meeting the challenge. Annual Review of Public Health, 14, 515–43.
Granger, C.W. (1998). The emerging science of functional assessment: our tool for outcome analysis. Archives of Physical Medicine and Rehabilitation, 79, 235–40.
Graves, E. (1999). National Hospital Discharge Survey: annual summary, 1996. Vital and Health Statistics, 13, 1–46.
Haddon, W., Jr (1963). A note concerning accident theory and research with special reference to motor-vehicle accidents. Annals of the New York Academy of Sciences, 107, 635–46.
Haddon, W., Jr (1972). A logical framework for categorizing highway safety phenomena and activity. Journal of Trauma, 12, 193–207.
Haddon, W., Jr, Schuman, E., and Klein, D. (1964). Accident research: methods and approaches. Harper and Row, New York.
Hedlund, J. (1985). Casualty reductions resulting from safety belt use laws. Presented at OECD Meeting, May, Washington, DC.
Hoyert, D.L., Kochanek, K.D., and Murphy, S.L. (1999). Deaths: final data for 1997. National Vital Statistics Reports, 47, 1–104.
Institute of Medicine (1997). Enabling America: assessing the role of rehabilitation science and engineering. National Academy Press, Washington, DC.
Institute of Medicine (1999). Reducing the burden of injury: advancing prevention and treatment. National Academy Press, Washington, DC.
Kahane, C.J. (1986). An evaluation of child passenger safety: the effectiveness and benefits of safety seats. National Highway Traffic Safety Administration, Washington, DC.
Kraus, J.F. (1993). Epidemiology of head injury. In Head injury (3rd edn) (ed. P.R. Cooper), pp. 1–25. Williams and Wilkins, Baltimore, MD.
Kraus, J.F., Peek, C., McArthur, D., and Williams, A. (1994). The effects of the 1992 California Mandatory Motorcycle Helmet Use law on motorcycle crash fatalities and injuries. Journal of the American Medical Association, 272, 1506–11.
Kraus, J.F., Peek, C., Silberman, T., and Anderson, C. (1995). The accuracy of death certificates in identifying work-related fatal injuries. American Journal of Epidemiology, 141, 973–9.
Langley, J. (1995). Experiences using New Zealand’s Hospital Based Surveillance System for injury prevention research. Proceedings of the International Collaborative Effort on Injury Statistics, Vol I, Chapter 9, pp. 1–8. Centers for Disease Control and Prevention, Department of Health and Human Services, Hyattsville, MD.
McCaig, L.F. and Stussman, B.J. (1997). National Hospital Ambulatory Medical Care Survey: 1996 emergency department summary. Advance Data, 293, 1–20.
McClure, R. (1995). Injury and general practice in Australia. Australian Family Physician, 24, 2059–63.
Mackay, M. (1985). Seat belt use under voluntary and mandatory conditions and its effect on casualties. In Human behavior and traffic safety (ed. L. Evan and R.C. Schwing), pp. 259–78. Plenum Press, New York.
MacKenzie, E.J., Damiano, A., Miller, T., and Luchter, S. (1996). The development of the functional capacity index. Journal of Trauma, 41, 799–807.
Miller, T.R., Lestina, D.C., and Galbraith, M.S. (1994). Medical care spending—United States. Morbidity and Mortality Weekly Report, 43, 581–6.
Murray, C.J.L. and Lopez, A.D. (1997a). Mortality by cause for eight regions of the world: Global Burden of Disease Study. Lancet, 349, 1269–76.
Murray, C.J.L. and Lopez, A.D. (1997b). Global mortality, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet, 349, 1436–41.
Murray, C.J.L. and Lopez, A.D. (1997c). Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet, 349, 1498–1504.
National Center for Health Statistics (1991). Impairments due to injuries: United States, 1985–87. Vital and Health Statistics, 10, 1–55.
National Committee for Injury Prevention and Control (1989). Injury prevention: meeting the challenge. American Journal of Preventive Medicine, 5 (Supplement).
National Highway Traffic Safety Administration (1984). Final regulatory impact analysis: amendment to FMVSS 208 passenger car front seat occupant protection. Department of Transportation, Washington, DC.
National Highway Traffic Safety Administration. (1992). Economic costs of motor vehicle crashes 1990. Department of Transportation, Washington, DC.
National Research Council (1985). Injury in America. National Academy Press, Washington, DC.
National Safety Council (1993). Understanding and preventing violence (ed. A.J. Reiss Jr and J.A. Roth). National Academy Press, Washington DC.
National Safety Council (1998). Accident facts. National Safety Council, Itasca, IL.
Nourjah, P. (1999). National Hospital Ambulatory Medical Care Survey: 1997 emergency department summary. Advance Data, 304, 1–23.
O’Carroll, P.W., Rosenberg, M.L., and Mercy, J.A. (1991). Suicide. In Violence in America: a public health approach (ed. M.L. Rosenberg and M.A. Fenley), pp. 184–96. Oxford University Press, New York.
Partyka, S.C. and Womble, K.B. (1989). Projected lives saved from greater belt use. In National Center for Statistics and Analysis Research Notes. National Highway Traffic Safety Administration, Washington, DC.
Peters, K.D., Kochanek, K.D., and Murphy, S.L. (1998). Deaths: final data for 1996. National Vital Statistics Reports, 47, 1–100.
Pollock, D.A. and McClain, P.W. (1989). Trauma registries: current status and future prospects, 1989. Journal of the American Medical Association, 262, 2280–3.
Powell, E.C., Sheehan, K.M, and Christoffel, K.K. (1996). Firearm violence among youth: public health strategies for prevention. Annals of Emergency Medicine, 28, 204–12.
Rennison, C.M. (1999). Criminal victimization 1998: changes 1997–98 with trends 1993–98. Bureau of Justice Statistics National Crime Victimization Survey NCJ 176 353, US Department of Justice, Washington, DC.
Rice, D., MacKenzie, E., et al. (1989). Cost of injury in the United States: a report to Congress, 1989. Institute for Health and Aging, University of California, San Francisco, and Injury Prevention Center, Johns Hopkins University, Baltimore, MD.
Rivara, F.P. (1999). Pediatric injury control in 1999: where do we go from here? Pediatrics, 103, 883–8.
Sherman, G.J. (1995). Federal injury surveillance in Canada: filling the gaps. Proceedings of the International Collaborative Effort on Injury Statistics, Vol I, Chapter 10, pp. 1–5. Centers for Disease Control and Prevention, Department of Health and Human Services, Washington, DC.
Sherman, L.W., Gottfredson, D., MacKenzie, D., Eck, J., Reuter, P., and Bushway, S. (1997). Preventing crime: what works, what doesn’t, what’s promising. National Institute of Justice Report 165 366, US Department of Justice, Washington, DC.
Stambrook, M., Moore, A.D., Peters, L.C., Devianene, C., and Hawryluc, G.A. (1990). Effects of mild, moderate and severe closed head injury on long-term vocational status. Brain Injury, 4, 183–90.
Tengs, T.O., Adams, M.E., Pliskin, J.S., et al. (1995). Five-hundred life-saving interventions and their cost-effectiveness. Risk Analysis, 15, 369–90.
United States Department of Health and Human Services (1995). Healthy People 2000, midcourse review and 1995 revisions. US Department of Health and Human Services, Washington, DC.
United States Department of Health and Human Services (1998). Healthy People 2010 objectives: draft for public comment. US Department of Health and Human Services, Washington, DC.
Waller, J.A. (1985). Injury control: a guide to the causes and prevention of trauma. Lexington Books, Lexington, MA.
Woodwell, D.A. (1997). National Ambulatory Medical Care Survey: 1996 summary. Advance Data, 295, 1–25.
Wright, M.A., Wintemute, G.J., and Rivara, F.P. (1999). Effectiveness of denial of handgun purchase to persons believed to be at high risk for firearm violence. American Journal of Public Health, 89, 88–90.
Zador, P.L. and Ciccone, M.A. (1993). Automobile driver fatalities in frontal impacts: air bags compared with manual belts. American Journal Public Health, 83, 661–6.