Introduction

In most developed nations unintentional injury is the leading cause of preventable death in children under 19 years.1–3 Elementary school children and teens are typically injured outside the home when making decisions about risk behaviours without adult supervisors, with many of these injuries involving play and recreational sports.4 Skateboarding is a sport that is gaining in popularity.5 6 With increasing growth of the sport, there has been a dramatic increase in the popularity of skateparks.7 Nonetheless, skateboarding is an activity that involves considerable risk of falling and as more skateparks are being built, injuries are emerging as a more pressing issue.7 8 The most common injury mechanism is losing one’s balance and falling off the skateboard,9 10 usually as a result of failing to successfully execute a trick.9 11 However, there is little known about the features on skateparks and types of tricks that increase injury risk. The current study addresses this gap by identifying the features preferred and the tricks youth engage in, as well as the frequency and severity of falls.

Medically treated injuries from skateboarding are particularly common for older children and adolescent skateboarders,8 12 13 possibly because of the types of tricks older youth attempt.14 In this study, therefore, younger (11–15 years) and older (16–20 years) skateboarders were unobtrusively video-recorded when on a large skatepark, with age group and sex estimated by physical features and dress.

MethodsParticipants

Skateboarders (n=526, 96% male) were assigned to one of two age groups (younger, n=166, 11–15 years; older, n=360, 16–20 years) based on visual assessment of physical characteristics (ie, height and facial features15). Inter-rater reliability15 was 92% agreement (kappa=90%) based on 25% of the videos; the data of the primary coder were analysed.

Materials

A tripod and digital video camera were unobtrusively located and recorded youth on the skatepark. Videos were coded using Noldus Observer XT software.

Data gathering procedure

The Silvercreek Skatepark (73 m long × 19 m wide, with a surface area of 1387 m2) comprises features common to many skateparks (eg, ramp, stairs, grind box) and was divided into seven smaller observable zones that were randomised during each observation day, with the same number of minutes spent observing in each zone.

Video coding

First, visual assessments were conducted to categorise each skateboarder into either the younger or older group, and as male or female (92% agreement, kappa=90%). Then the behaviours of each skateboarder within a zone were coded for feature used (100% agreement between two independent coders based on 25% of the videos) and tricks attempted (94% agreement, kappa=92%); see the online supplemental table for details about skatepark features and videos showing common tricks.

A fall severity scale was applied to code anytime youth attempted a trick or fell from their skateboard: 1=skateboarder completed the trick successfully; 2=skateboarder swayed or shuffled the skateboard due to a loss of balance; 3=skateboarder put one foot off of the skateboard; 4=skateboarder put both feet off; 5=skateboarder fell onto part of their body; 6=skateboarder fell and onto their entire body; inter-rater reliability for 25% of videos was 93% agreement (kappa=91%). The data of the primary coder were analysed. Note that for the flat ground feature, usage was not coded when locomoting from one feature to another on the ground. We had intended to also code helmet usage, but this occurred too infrequently to track.

Data preparation and analysisPart 1

The proportion of children using each feature was computed for each age group, as was the proportion of falls on each feature. An injury risk exposure score was calculated (proportion of use × proportion of falls), with higher numbers indicating greater fall risk on a given feature.

Tricks were grouped based on movement similarities. The proportion of children in each age group that attempted each type of trick was computed, as was the proportion of falls for each type of trick. The 95% CIs for feature means were calculated via SPSS’s ‘explore’ function. Data are available on request.

Part 2

Data were collapsed across age groups and analysed using SPSS V.26. Analyses included paired Wilcoxon signed-rank tests (ie, data were overdispersed and severity scores were skewed) to compare the success score (max=1.0) and the fall severity scores associated with different features and tricks to the most popular reference group (feature: ground, trick: jump).

Patient and public involvement

There was no patient or public involvement to report for this research.

Discussion

Skateparks can be a sociable place for youth to exercise and participate in activities they enjoy.7 16 17 However, they also contain a combination of concrete and metal features with surfaces that can cause significant injury when falls occur. The present study addressed notable research gaps relevant to skatepark design and the tricks that youth attempt on the skatepark that increase risk of falls.

In terms of skatepark design, there are no universal guidelines or safety standards and we located only three reports that are relevant. One focused on how to limit flaws (https://publicskateparkguide.org/design-and-construction/factors-of-skatepark-design/). The two other reports were written from an insurance perspective and considered risk management and accident reporting (https://www.emcins.com/losscontrol/techsheet/skate-park-design-and-safety; https://www.emiia.org/files/miia_skatepark_guidelines_2008_edition.pdf). The present findings suggest a number of conclusions about features that are relevant to the design of skateparks. The flat ground was the most popular area at both ages, along with the quarter pipe and ramp features. The pattern of feature usage by age, however, suggests that the grind box is more popular with older experienced skateboarders than younger ones. A differential pattern of feature usage by age could be reflected in how skateparks are designed. For example, features might be grouped together strategically to encourage younger skateboarders in one region and older skateboarders in another. Segregating ages by design might reduce imitation of risky tricks performed by older skateboarders that are viewed by younger skateboarders but are beyond their skill set. Youth imitating peers who are modelling risk behaviours during play is a risk factor for injury.18–21

With regard to designing for safety, at both ages, a majority of skateboarders (about 72%) used the flat ground and fall outcomes occurred at comparably high instances (about 63% of youth fell). Although the flat ground poses no inherent risk due to design characteristics, skateboarders were obviously able to create risk by attempting tricks, including flips, jumps and turns. Skateboarders’ desire to master new tricks that challenge their capabilities may contribute to explain this finding.22 For youth at both ages, the quarter pipe and ramp also were popular, though to a lesser degree (45–51% of youth), and fall outcomes on these features did not vary by age, with approximately 37% of skateboarders falling. Additionally, approximately 35% of older youth also used the grind box, with 67% falling; young skateboarders also had a high frequency of falls on the grind box (85% fell) though not many used this feature (14%). It may prove useful to moderate risk of falling if features are differentially coded across the skatepark to suggest difficulty levels, much like what is done on ski hills. In sum, three features were popular at both ages and associated with high to moderate frequency of falls: the flat ground, the quarter pipe and the ramp. The grind box also was a high-risk feature for falls at both ages, though it was more often used by older than younger skateboarders, suggesting some degree of experience and skill level was judged to be needed before a skateboarder attempted use of that feature. Applying these findings to the design of skateparks, the results suggest that some features are seldom used (stairs, railings), others are popular at all ages (flat ground, quarter pipe, ramp) and some are more popular for more experienced older skateboarders (grind box).

Falls occur on virtually all features (see table 1). However, youth generally fall more frequently on some features (flat ground, grind box) than others (quarter pipe, ramp), with certain features posing differential risk of falls for older skateboarders (railing). Suffice it to say, limiting to those features used by at least 10% of skateboarders herein (table 1) suggests that if a skatepark includes both a quarter pipe and ramp feature then there will likely be a lower risk of falls than if there is a grind box and flat ground as the primary features. Including railings and stairs appeals more to older than younger skateboarders but these are high-risk features for falling by these users. Applying these findings in the design of skateparks could be an effective approach to moderate risk of fall injuries from skateboarding.

With regard to tricks, at both ages youth tended to favour the same tricks, and both slide and flip tricks were high-risk tricks that resulted in an elevated frequency of falling (67–94% of youth fell). The fact that both younger and older youth showed the same rank ordering of fall frequency as a function of trick type suggests that experience skateboarding may have limited impact to reduce fall risk. Rather, what may matter is the type of trick attempted, with some types elevating fall risk more than others. Experience with a recreational activity often leads youth to engage in more risk-taking behaviours, which essentially counteracts any possible reductions in injury risk that might accrue with improvements in skill level from accumulated experience.23 It may be, therefore, that reducing skateboarders’ risk of injuries resulting from falls during these high-risk tricks may require designing skateparks in ways to minimise the occurrence of these tricks (environment modification) and to identify ways to increase usage of personal protective gear (individual behaviour change). Based on our observations of so few youth using helmets, which is consistent with a recent historic review on this topic,24 motivating usage is likely going to require a shift in attitude and/or mandating usage with rigorous enforcement to support that behaviour change.

Collapsing over age and examining both the success and severity outcomes of tricks for all youth supports several conclusions (table 4). Flip tricks were less successfully executed and resulted in falls that were more severe. Designing skateparks with features that limit these will likely reduce fall injuries. In contrast, turn tricks were more often successfully executed and the falls were less severe. Thus, designing parks in ways that supported the execution of turn tricks may contribute to fun experiences while limiting risk of severe falls. The category of ‘other’ types of tricks was often successful; however, when falls occurred they were more severe. Our findings revealed a variety of tricks under ‘other’ and these were most popular for older skateboarders. A more thorough examination of what features are implicated in the execution of these tricks that pose risk for serious falls is an important next step in order to determine if there are other features that might be avoided to reduce the severity of fall injuries on skateparks.

Limitations and future research directions

Although reliability for child age and sex was excellent, validating these estimates by directly gathering demographic information from youth is recommended. Additionally, as expected our sample was primarily male,8 12 13 25 but it would be useful to strategically recruit female skateboarders and assess if their results differ, since there are differences in size and strength of females and males. Finally, this study measured fall severity based on what was observed and the potential for injury. It would be useful to talk directly with skateboarders who have experienced a medically attended injury and assess how these affect their decisions with regard to features they will use and/or tricks they plan to attempt on returning to the skatepark.