To describe the circumstances, severity, and outcomes of skating-related injuries among children admitted to trauma centers.
A cross-sectional comparison of roller skaters (n = 154), in-line skaters (n = 190), and skateboarders (n = 254) aged 5 to 19 years who were hospitalized with injuries.
Seventy-nine hospitals and pediatric trauma centers participating in the National Pediatric Trauma Registry between October 1988 and April 1997.
Three quarters (75.8%) of the study sample were male, nearly half (47.8%) were injured on roads, and more than one third (37.1%) had head injuries. Among skateboarders, 50.8% had head injuries compared with 33.7% of in-line skaters and 18.8% of roller skaters (P<.001). According to the Injury Severity Score, injuries to skateboarders were 8 times more likely to be severe or critical compared with roller skaters' injuries and more than 2 times as likely to be severe or critical compared with in-line skaters' injuries. Mean hospital length of stay was 6.0 days for skateboarders, 3.4 days for in-line skaters, and 2.4 days for roller skaters (P<.001). Skateboarders were more likely to be male and to be injured on roads than were in-line skaters or roller skaters.
Skateboarding-related injuries are more severe and have more serious consequences than roller skating or in-line skating injuries. Research is needed to identify ergonomic and behavioral factors responsible for higher head injury risk to skateboarders, and interventions are needed to reduce the risk.
WAKING AFTER a long sleep, a modern-day Rip Van Winkle would be amazed at the skaters and skateboarders whizzing by on streets, sidewalks, and bicycle paths. This is a new kind of traffic—often barely in control, faster than a pedestrian, but slower than a bicycle or motor vehicle. Skating has become a major recreational activity, and it is emerging as a new mode of urban transportation.1 Injuries associated with the use of roller skates, in-line skates, and skateboards have skyrocketed accordingly.2,3
Most studies of skating injuries are based on National Electronic Injury Surveillance System (NEISS) data, which is maintained by the Consumer Product Safety Commission.4 According to this national database, approximately 160000 roller skaters, skateboarders, and in-line skaters were treated in hospital emergency departments (EDs) between July 1992 and June 1993. The male-female ratios were very different for the 3 types of injured skaters.5 Nearly all skateboarders were male, as were most in-line skaters; however, most roller skaters were female. The median ages were 12 years for roller skaters, 13 years for skateboarders, and 15 years for in-line skaters.
Most in-line skating injuries are caused by forward falls on outstretched arms, without vehicle, bicycle, or other-skater involvement.6 It follows that articles describing skating injuries often focus on wrist and upper extremity injuries.7,8 However, lower extremity injuries5,9 and head injuries also occur and can be serious.5
Results of research on skateboard-related injuries suggest that head injuries are more common among younger children and that extremity injuries are more common among older children.10,11 However, because older skateboarders tend to skate faster and on "streets and highways," when they sustain head injuries, they are more severe.10 This illustrates the complexity of research in this area: different types of skating appeal to males and females of different ages. In-line skating attracts the broadest age range, skateboarding is done primarily by males aged 11 to 14 years, and most roller skaters are younger females. In addition, type of skate seems to be related to skating location (skate park vs city street) and to use of protective gear. In turn, skating location and use of protective gear affect the type and severity of injuries.3,7
One trend is clear. Without major changes in skating safety behavior, skating-related injuries and deaths are expected to rise.7,12 A high proportion of injuries occur to novice skaters3 and skateboarders,13,14 suggesting that many skating-related falls are preventable if skaters take the time to learn the basics while skating on flat, smooth, dry surfaces.6,9,15 Moreover, when falls do occur, injuries can be prevented or minimized by wearing appropriate protective gear.3,7,16
Most research on skating injuries has examined the early acute phase—what happens in the physician's office or in the ED. Follow-up treatment and services and hospital care have rarely been studied. To gain a more accurate picture of skating-related injuries, we examined the index hospitalization of children admitted for roller skate, skateboard, and in-line skate injuries.
The data for this study were extracted from the National Pediatric Trauma Registry (NPTR).17 The NPTR is a voluntary multi-institutional database with information on many aspects of pediatric trauma. At the time these data were abstracted, 79 children's hospitals or pediatric trauma centers contributed data to the NPTR.
A trained trauma nurse coordinator at the participating institution completes a data collection form for each trauma patient according to instructions developed by the NPTR and explained in an operational definitions manual.18 To guarantee uniformity across institutions, coding for natural and external causes of injury, severity scoring, data management, analyses, and reporting are performed centrally by the staff at the NPTR.
The NPTR includes children and adolescents from birth to age 19 years who are admitted to the hospital for an acute injury, including patients who are dead on arrival or who die in the ED. All injuries are included except burning, poisoning, and near drowning. From October 1988 to April 1997 there were 62190 cases recorded in the registry.
After selecting sports-related injury as the mechanism of injury, relevant NPTR cases were detected via a word search of the injury description. The search identified children injured while skating and skateboarding. We excluded ice skaters and skaters of unspecified type. In addition, because of the rarity of skating injuries in young children, 6 children younger than 5 years were excluded.
The final sample included 598 cases involving roller skates, in-line skates, and skateboards consecutively recorded in the NPTR between October 1988 and April 1997, including 8 children who died, 4 of whom were skateboarders. Each child was admitted to the hospital between October 1988 and January 1997.
Based on findings in the entire sample, we then analyzed 2 subsamples. The sample of 598 children was first restricted to male in-line skaters and skateboarders who were injured in falls, regardless of place of occurrence (n = 284), and then to the smaller subset of males injured in falls on the road (n = 140). Vehicle-related events were excluded from the 2 subsets to allow us to examine the specific injuries associated with each type of skate. Roller skaters were also excluded because their demographic profile was different and they tended to skate in different locations.
We analyzed the type of skater by year of injury, sex, age, injury setting, causes of injury, number of diagnoses, anatomical region of injury, injury severity, functional status, and index hospitalization. Severity of injury was measured by the Injury Severity Score (ISS).19 In our analysis, we grouped ISS values into the following categories: trivial, 0 to 3 (0 indicates no injury); minor, 4 to 8; moderate, 9 to 15; severe, 16 to 24; and critical, 25 to 75. Functional status was assessed at discharge by rating the child's ability in 9 functional domains: vision, hearing, speech, self-feeding, bathing, dressing, walking, cognition, and behavior. The child's performance in these functional areas was rated by a clinician at each of the participating trauma centers as being age appropriate, impaired, or unable, and we combined the last 2 categories.
Cross-tabulations with χ2 and difference of means tests were calculated using a statistical analysis software program (Statistical Package for the Social Sciences [SPSS] for Windows, SPSS Inc, Chicago, Ill).20 Graphs were made using a software program (Quattro Pro for Windows, Borland International Inc, Scotts Valley, Calif).21
Figure 1 shows that the type of skater in our sample changed drastically from 1989 to 1996. No in-line skaters were even reported to the registry in the first 2 years, but by 1996 they accounted for 55% of all skating-related injuries in the NPTR sample. On the other hand, skateboarders made up approximately three quarters (76%) of the cases in 1989 but dropped to one quarter (25%) by 1996.
Type of skater, by year.
Table 1 shows a breakdown of key variables by type of skate—roller skate (n = 154), in-line skate (n = 190), and skateboard (n = 254). There were some fundamental differences in the demographic profiles of children using these 3 types of skates, the settings in which they were injured, and the causes and severity of their injuries. In fact, all variables were significantly related to the type of skate (P<.001).
In this series, 93.3% of skateboarders were male, 82.6% of in-line skaters were male, and only 38.3% of roller skaters were male. Skateboarders were the oldest, followed by in-line skaters, and then roller skaters.
Table 1 shows that, in this sample, children injured on skateboards had more serious injuries than the other 2 types of skaters. Based on the ISS, only 2.0% of the children injured while roller skating had severe or critical injuries compared with 7.4% of in-line skaters and 16.5% of skateboarders. The average number of diagnoses was 1.3, 2.0, and 2.3, respectively. Similarly, mean hospital length of stay was 2.4 days for roller skaters, 3.4 days for in-line skaters, and 6.0 days for skateboarders, or a total of 2531 acute care hospital days.
Only 14.9% of roller skaters sustained their injuries on the road compared with 54.7% of in-line skaters and 62.6% of skateboarders (P<.001). Similarly, only 2.6% of roller skaters were hit by vehicles compared with 22.1% of in-line skaters and 24.8% of skateboarders (Table 1).
Figure 2 shows that skateboarders were more likely to have head injuries (50.8%) than were in-line skaters (33.7%) or roller skaters (18.8%). Roller and in-line skaters were significantly more likely to have upper extremity injuries compared with skateboarders. The proportion with lower extremity injuries was similar for all 3 types of skaters, and the differences were not statistically significant.
Type of skater, by anatomical region of injury, based on the full sample (N = 598). Note that percentages do not sum to 100 because many children had injuries to several body regions.
Table 2 shows injury-related impairments at discharge from the hospital. Vision, hearing, speech, behavior, and cognition impairments were rare, which is fortunate because these types of impairments can be serious and long-lasting. In contrast, self-feeding, walking, dressing, and bathing impairments were common; fortunately, these impairments are not usually as serious or long-lasting.
To understand the impact of injured body region, we looked at children with injuries to only 1 body region: head (n = 185), upper extremity (n = 206), or lower extremity (n = 129). Based on the ISS, children with upper and lower extremity injuries had less severe injuries than children with head injuries (P<.001). Looking at the same groups of children, the mean index hospitalization was 1.8 days for children with upper extremity injuries compared with 4.4 days for children with head injuries and 5.8 days for children with lower extremity injuries. When looking at the sum of acute care hospital days in the 3 groups, upper extremity injuries account for fewer acute hospital days (19%) than either head injuries (42%) or lower extremity injuries (39%).
Finally, in the full sample (N = 598), cross-tabulations of sex by ISS (in 5 categories) show that males had more severe injuries than females (P<.01). Males were more likely to be injured on the road than were females (66% vs 30%; P<.001), and 42% of males had head injuries vs only 23% of females (P<.001).
The inquiry now turns to exploring the differences between in-line skaters and skateboarders injured in falls. To remove potential confounding variables (sex and cause of injury), we restricted these analyses to males injured in falls on skateboards or in-line skates. In other words, we excluded females, roller skaters, and children who were hit by motor vehicles.
In this subsample, large differences remained between the 174 male skateboarders and the 110 male in-line skaters injured in falls. For example, skateboarders were still significantly older than injured in-line skaters (mean age, 12.8 vs 11.7 years; P<.01), and they tended to have more severe injuries (P<.05). Figure 3 shows the anatomical regions of injury for the 2 groups. Compared with in-line skaters, skateboarders were significantly more likely to have lower extremity injuries (21.8% vs 10.9%) and head injuries (46.0% vs 19.1%) and significantly less likely to have upper extremity injuries (32.8% vs 63.6%).
Type of skater, by anatomical region of injury, based on a subsample of males injured in falls (n=284).
Although children hit by motor vehicles were excluded, place of injury may remain a confounding variable; 66.9% of this subsample of skateboarders were injured on roads compared with 54.0% of in-line skaters (P=.06). Consequently, the sample was further restricted by including only falls that occurred on roads.
After restricting the sample to males injured in falls on roads, 140 cases remain. In-line skaters still tended to be younger than skateboarders (mean age, 11.9 vs 13.0 years; P <.05) and to have less severe injuries (P<.05). Previous differences between in-line skaters and skateboarders in percentages with lower extremity injuries disappeared. However, the 2 remaining differences were highly significant: 49.5% of skateboarders had head injuries vs only 23.4% of in-line skaters, and 31.2% of skateboarders had upper extremity injuries vs 51.1% of in-line skaters (Figure 4).
Type of skater, by anatomical region of injury, based on a subsample of males injured in falls on the road (n=140).
Three major findings emerge from this study: (1) skateboarders have more serious injuries and are more likely to have head injuries than in-line skaters or roller skaters, (2) males are more likely to be injured skating than females and to have more serious injuries, and (3) upper extremity injuries are most salient in ED-based studies of skaters, but in our hospital-based sample, head and lower extremity injuries seem to be more prominent.
In our sample, skateboarders had more serious injuries than in-line skaters, and roller skaters had the least severe injuries. These findings contrast sharply with those of Schieber et al,5 who report that 52% of in-line skaters are in the "more severe" category compared with 46% of roller skaters and only 36% of skateboarders.
These 2 studies differ in 3 fundamental ways, so we did not expect concurrence. First, the sample studied by Schieber and colleagues was based on ED visits and ours was based on hospital admissions. In fact, only 27% of the children in our sample were treated in an ED before arriving at the hospital. Second, Schieber and colleagues used their own severity measure, which was calibrated differently than the ISS, which we used. For example, a child whose only injury is a concussion with brief loss of consciousness is considered "more severe" on their measure but has only a "minor" injury according to our ISS categories. Third, NEISS only records the most serious injury diagnosis, and as many as 15 diagnoses are recorded in the NPTR.
In our study, much of the difference in severity across skater types seems to be related to differences in age, sex, skating location, and motor vehicle involvement. Skateboarders seem to have more serious injuries than in-line skaters partly because they tend to be older (adolescent) males who skate on roads and are hit by motor vehicles. However, even after restricting the sample to males injured in nonvehicular falls on the road, there were some major differences between skateboarders and in-line skaters. Skateboarders were still older, had higher ISSs, were twice as likely to have head injuries, and were much less likely to have upper extremity injuries (Figure 4).
The finding that skateboarders' injuries were more serious than those of roller skaters is consistent with findings by Baker and colleagues.2 Their analysis looked at children treated in EDs after being injured in a variety of recreational activities (on playgrounds, on roller skates, on skateboards, or riding on children's vehicles). Skateboarding was the mechanism of injury most likely to result in a hospital admission; almost 13% of children seen in the ED for skateboarding injuries were admitted to the hospital.
The preponderance of head injuries among skateboarders merits further examination. Future research on roller and in-line skaters and skateboarders is needed to explore differences in skating safety behavior and ergonomics. Were skateboarders in this sample more prone to head injuries because they engaged in different kinds of skating (eg, aggressive stunt skating), skated at dangerous locations (eg, on streets), had different types of falls (eg, backward vs forward), had less control over their skateboards, or were less likely to wear helmets?
This sample was disproportionately male, and males had more serious injuries than females. Results of previous research show that males are at highest risk of skating-related injuries because they do more outdoor skating, they skate in more dangerous locations, and they wear less protective gear than females.1,22 Roller skaters are different, however. In our sample, more than 60% of roller skaters were female, and they were significantly younger than the other types of skaters. In addition, two thirds of their injuries occurred at home or in recreation areas, where speeds are slower and encounters with motor vehicles are rare.
These age and sex distributions are similar to the ED-based data of the NEISS.4 In the NEISS data, however, roller skaters are grouped with ice skaters and other unspecified types of skaters. Although this makes comparison difficult, roller skaters still seem to be disproportionately female and younger than other types of skaters.
The percentage of children having upper extremity injuries in this sample is much lower than the percentages reported by other researchers for in-line skaters,5,7- 9 roller skaters,23,24 and skateboarders (Table 3).5 One reason is that the other studies are based on ED data and our study is based on hospital data. By themselves, most upper extremity injuries are not severe enough to require inpatient hospitalization.
Among children in the NPTR sample, body region of injury had a profound impact on severity and outcomes. Children with head injuries and injuries to the lower extremities had more severe injuries (based on the ISS) and spent significantly more days in the hospital than children with upper extremity injuries.
Altogether, these 520 children accounted for 1928 days in the hospital. Of those days, children with head injuries accounted for 42%, children with lower extremity injuries accounted for 39%, and children with upper extremity injuries accounted for only 19%.
Comparing skateboarders with roller skaters and in-line skaters provides a context for understanding skateboarding and 2 alternatives to skateboarding. Based on these results, would-be skaters should be encouraged to take up roller skating or in-line skating instead of skateboarding.
Only 7% of skateboarders were female compared with 62% of roller skaters. It follows that different prevention strategies may be needed for skateboarders and roller skaters, just as they are for in-line skaters and roller skaters.25 For instance, given the preponderance of males being injured on skateboards, clinicians and health educators may want to target males for prevention efforts.
The high percentage of in-line skaters wearing wrist guards1,26,27 suggests that in-line skaters are aware of the danger of wrist and forearm injuries. Although some emphasis should remain on upper extremity injury, more attention should now be given to preventing skating-related head and lower extremity injuries. Although head and lower extremity injuries are less common than upper extremity injuries, they deserve more attention because they are generally more serious and require a longer hospitalization period.
Many of the head injuries in our sample would have been prevented if the children had been wearing helmets. Along with knee pads, the best way to prevent lower extremity injuries is to avoid the crash altogether. Better training and skating at safer locations can help skaters avoid bad falls and can reduce the chances of getting hit by a motor vehicle.
Increasingly, athletes are using in-line skates for cross-training.6,7 Thus, along with "garden variety" recreational skaters, clinicians are likely to encounter growing numbers of young athletes, as well as student and work commuters.9,12,13
Clinicians can help prevent skating-related injuries in a number of ways: by communicating the injury risk to children and parents (especially risks associated with skateboarding); by advising patients who skate to wear protective gear (helmets, knee and elbow pads, and wrist guards); by advocating for safe skating areas (eg, skate parks, multipurpose paths, and skate-friendly pedestrian ramps); by strongly recommending that skaters take lessons; by doing postinjury counseling about proper skating safety behavior; or by championing effective approaches for reducing skating injuries, including legislation.
We went one step beyond previous research that examined ED cases.6,7,9,12 We examined circumstances, severity, and outcomes in the subset of children with serious skating-related injuries requiring hospital admission. Only 27% of the children in our hospital-based sample even went to an ED before hospital admission. In other words, there is not much overlap between our hospital-based sample and previous samples of ED patients. An important next step is to collect data on the even smaller subset of skaters who are discharged from the hospital with functional limitations. We expect that this would lend further support to our emphasis on less common but more serious injuries to the head and lower extremities.
The ISS can serve as a common denominator, allowing clearer comparisons of skaters in different samples. In Table 1, we show the ISSs of children in this sample. Whether examining skaters in EDs or in hospitals or looking at the subset of those discharged from hospitals with impairments, future researchers are encouraged to collect and display similar ISS data.
Skating is a public activity, yet systematic data on this emerging sport are not yet available. Observational data are needed describing skating safety behavior and groups at risk for sustaining outdoor skating injuries. Physicians and other health care professionals (namely, nurses, physical and occupational therapists, casting professionals, and researchers) are urged to give more attention to patients with skating-related injuries. Clinical participation is needed in studies that document the types of injuries, treatments and services provided, and outcomes attained.
Accepted for publication May 4, 1998.
This work was supported by grant H133B50006 from the National Institute on Disability and Rehabilitation Research, US Department of Education, Washington, DC, and grant R491CCR302486 from the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Ga.
Many thanks to the physicians and trauma registry coordinators at the National Pediatric Trauma Registry participating hospitals. In addition, we thank Sarah C. Stiles, PhD, JD, for reading and commenting on several versions of the manuscript.
Reprints: J. Scott Osberg, PhD, New England Medical Center, 75 Kneeland St, Fifth Floor, Rehabilitation, Boston, MA 02111 (e-mail: email@example.com).
Editor's Note: And you thought skating on thin ice was dangerous!—Catherine D. DeAngelis, MD
Osberg JS, Schneps SE, Di Scala C, Li G. SkateboardingMore Dangerous Than Roller Skating or In-line Skating. Arch Pediatr Adolesc Med. 1998;152(10):985-991. doi:10.1001/archpedi.152.10.985