o RIGINAL RESEARCH_______________________________ Injuries Involving Off-road Cycling Frederick P. Rivara, MD, MPH; Diane C. Thompson, MS; Robert S. Thompson, MD; and Viviana Rebolledo Seattle, Washington BACKGROUND. Data on injuries due to off-road bicycling are scarce, but do indicate that injuries in this sport are frequent. We examined the pattern of injuries to off-road riders as part of a larger study of bicycle injuries and helmet use. METHODS. We undertook a prospective study of bicycle-related injuries identified at seven emergency depart­ ments in Seattle, Washington, between March 1992 and August 1994. Hospitalized patients and medical examin­ ers’ cases were included. Detailed questionnaires and abstraction of all medical records provided information on crash and rider characteristics and injury type and severity. RESULTS. A total of 3390 injured riders participated, representing an 88% response rate. Of all injured cyclists, 127 (3.7%) were injured riding “off road.” Seventy-three percent of off-road cyclists were 20 to 39 years of age, and 86.6% were male. Helmet use was 80.3% for off-road cyclists as compared with 49.5% for other cyclists. The number of head and face injuries for the off-road cyclists was only 40% of the number incurred by other cyclists. Four percent of off-road cyclists had severe injuries (Injury Severity Score >8), and 6.3% were hospital­ ized, compared with 6.8% and 9.4%, respectively, of other cyclists. CONCLUSIONS. The majority of off-road bicycling injuries are minor. Off-road cyclists are less likely to have head and face injuries than other cyclists and are more likely to wear helmets. KEYWORDS. Bicycling; head protective devices; facial injuries; head injuries. (J Fam Pract 1997; 44:481-485) ff-road bicycles (known as mountain frequent. Chow et al2 found that 83% o f competitive bikes) have become increasingly pop­ off-road cyclists and 51% of recreational off-road ular in the United States. These bikes cyclists reported injuries during the preceding year. are designed to ride on various types In another study, Kronisch and Rubin3 found that o f unpaved surfaces, such as hiking 20% of those they surveyed sustained significant trails, ski slopes, dirt roads, and other roughtraumatic ter­ injuries requiring medical attention and O limitations in activity during the preceding year. rain. Sales o f off-road bikes accounted for 62% o f the US bicycle market in 1992.' In 1983, the Because of the tremendous growth of off-road National Off-Road Bicycle Association (NORBA) biking and the potential importance o f the injuries was organized to promote the sport and organize incurred, we examined the pattern of injuries to off­ races. In 1996, off-road cycling will have a full road cyclists as part of a larger study of bicycle Olympic medal competition. injuries and helmet use. Data on injuries due to off-road biking are scarce, but do indicate that injuries in the sport are METHODS Submitted,, revised, January 16, 1997. D e s ig n From the Harborview Injury Prevention and Research Center, This study was part of a larger study of bicycle Department o f Pediatrics (F.P.R., D.C.T., V.R.), and injuries and helmet effectiveness.4 Data collected Department o f Epidemiology ( F.PR.) , University o f Washington; and the Department o f Preventive Care, Group as part o f this larger study gave us the opportunity Health Cooperative o f Puget Sound ( R.S. T.), Seattle, to examine a relatively large series of injuries to Washington. Requests fo r reprints should be addressed to off-road cyclists. Subjects were recruited from Frederick P. Rivara, MD, MPH, Harborview Injury Prevention arid Research Center, 325 N inth Ave, Box 359960, Seattle, WA seven Seattle area hospitals: Central and Eastside 98104-2499. Hospitals of Group Health Cooperative of Puget © 1997 Appleton & Lange/ISSN 0094-3509 The Journal o f Family Practice, Vol. 44, No. 5 (May), 1997 481 OFF-ROAD CYCLING INJURIES Sound, a large staff-model HMO; Harborview Information on the injuries was gathered from Medical Center, a Level I Regional Trauma Center; the emergency department, hospital, and medical University o f Washington Medical Center, a tertiary examiner’s records. These records were abstracted university hospital; Overlake Medical Center, a by trained abstractors, using a standardized form. community hospital serving the east side o f Seattle; The Abbreviated Injury Scale (AIS)5 was used to Children’s Hospital and Medical Center, a tertiary assess the severity of injuries, and an Injury children’s hospital in Seattle; and Mary Bridge Severity Score (ISS)6 was calculated. This was Children’s Hospital, a community and secondary done using a commercial computer program, TRI­ level children’s hospital in Tacoma, Washington. CODE,7 which converts textual injury descriptions The records of the King and Pierce counties med­ into International Classification of Diseases, 9th ical examiners’ offices were also examined during R evision (ICD-9) codes and AIS codes and calcu­ the study period to detect out-of-hospital deaths lates AIS scores for each injury and an overall ISS from bicycling. score. We used the TRI-CODE program to provide consistent and accurate injury coding and injury Subject Identification severity scoring, given that the data were abstract­ Injured cyclists were identified by regular surveil­ ed from seven different hospitals. Injuries with ISS lance of emergency department logs or treatment scores > 8 were defined as severe, and those with forms in each of the seven hospitals at least 1 to 2 ISS scores < 8 w ere nonsevere. times per week during the study period, March 1, 1992, through August 31,1994. Any person who was Data A nalysis injured while on a bicycle (whether moving or not) Data were double key-entered, cleaned, and ana­ was eligible for the study. Pedestrians who were lyzed using SAS software.8 Univariate analyses injured by a bicycle were not included. Only child were conducted to examine associations between passengers on bicycles were included. Also exclud­ type o f cycling and injury severity, crash character­ ed were cyclists sustaining injuries from assault istics, and demographic characteristics. Odds ratios while riding a bicycle. (OR) were used to express the likelihood of these findings in off-road cyclists, compared with other Data Collection riders unless otherwise indicated. Odds ratios and Detailed questionnaires were sent to all study sub­ 95% confidence limits were calculated using the jects. Those who did not respond were telephoned maximum likelihood method and the Cornfield approximately 14 days after the initial mailing. method, respectively. Unconditional logistic regres­ The questionnaires included inquiries about the sion was used to calculate odds ratios adjusted for demographic characteristics of the subjects, a number of covariates. cycling experience, circumstances of the crash, self-reported speed, severity of damage to the RESULTS bike, ownership and use of helmets, and self- reported helmet fit. Helmets were classified as During the study period, there were 3849 eligible hard shell if they were manufactured with a thick subjects treated in the emergency departments of plastic shell over a polystyrene liner; thin shell if the seven study hospitals. In addition, 5 subjects died the shell was very thin, flexible plastic over the from bicycle-related injuries before arriving at the liner; and no shell if there was no or only a cloth emergency department and were identified from the outer shell over the liner. medical examiner’s records. We obtained completed Since the majority of bicycles sold currently in the questionnaires and injury data on 3390 subjects for United States are off-road bikes, the type o f bicycle an overall response rate o f 88.0%. could not be used to identify subjects injured while There were 127 cyclists who reported that they riding off-road. Instead, we used the reported loca­ were injured while riding off-road, accounting for tion o f the crash to identify these subjects. This 3.7% of injured cyclists. Nearly three fourths of the included any cyclist reporting injury while riding off­ off-road injured cyclists were 20 to 39 years old. Tire road, off-road trails, forest or logging roads, and characteristics of the off-road cyclists were com­ forested or dirt areas o f city parks. pared with the 3263 bicyclists who were injured on 482 The Journal o f Family Practice, Vol. 44, No. 5 (May), 1997 OFF-ROAD CYCLING INJURIES TABLE 1 ------------------------------------------------------------------------------------------------------ helmets; there were no differ­ Characteristics of Persons Involved in Off-road Biking Injuries Compared with Persons ences in the proportion of Injured in Other Biking Crashes (N=3390) cyclists in the two groups using Off-road Cyclists Other Cyclists helmets without shells. Ap­ (n=127) (n=3263) proximately two of five helmets Characteristic No. (%) No. (%) P Value* in both groups of cyclists were visibly damaged in the crash. Sex .001 Overall, the body parts 110 (86.6) 2328 (71.3) Male injured (Table 3) and the types Female 17 (13.4) 9 3 5 (28.7) o f injuries sustained (Table 4) Age, y .001 were similar in the two groups 0 — 253 (7.8) <6 o f riders. There were, however, 6-12 4 (3.1) 1212 (37.1) 13-19 2 0 (15.7) 524 (16.1) some notable exceptions. Off­ 20-39 9 3 (73.2) 9 36 (28.7) road cyclists were about half as >40 10 (7 .9 ) 3 38 (10.4) likely to sustain head and face Frequency of ridingf .024 injuries (Table 3). They were Daily 63 (50.5) 2039 (62.7) less likely to sustain abrasions Weekly 48 (38.1) 8 36 (25.7) and be treated for dental Monthly 8 (6.3) 149 (4.6) <M onthly 2 (1.6) 72 (2.2) injuries than were other riders Seldom 6 (4.7) 157 (4.8) (Table 4).