RECREATIONAL VEHICLES (WORKING GROUP SESSION)

Transcription

1 RECREATIONAL VEHICLES (WORKING GROUP SESSION)

2

3 Fatal Recreational Vehicle Crashes and Alcohol in New Zealand: What does the data tell us? D.J.Begg Injury Prevention Research Unit, Dunedin School of Medicine, University of Otago, P O Box 913, Dunedin, New Zealand. Keywords ATVs, bicycles, recreational vehicles, alcohol Abstract Previous research in New Zealand has examined all non-traffic motorcycle/atv crash fatalities, bicyclist injuries, and the involvement of alcohol in drownings. No attempt has been made, however, to provide an overall picture of fatalities associated with recreational vehicles, and to determine the extent to which alcohol may have been involved in these events. The aim of this study was to examine existing electronic databases to identify fatal crashes involving the operator of a recreational vehicle, and determine the extent to which the involvement of alcohol could ascertained from these data. The New Zealand Health Information Services (NZHIS) fatality database includes records of all fatalities that occurred in New Zealand. For the purposes of this study, cases for (inclusive) with the following E-codes (external cause of injury codes) were selected: all non-traffic motor vehicle crashes, excluding pedestrians (E ); bicyclist crashes both traffic and non-traffic (E with 4 th digit 6 and 826 with 4 th digit 1; other road vehicle crashes 829; on-road ATV operators (E th digit 2 and 3). For each of these records the free text event description was manually searched to verify the involvement of a recreational vehicle, and any mention of alcohol involvement. Over the five year period 115 cases were identified as recreational vehicle fatalities, 80 (69%) of which were bicyclists, 21 involved 4-wheel motorcycle, 4 involved other motorcycles The victims were predominantly male, and 50% were aged less than 24years. Alcohol was recorded for only 6 cases, 5 of which were positive blood alcohol concentration (BACs). From this it is not possible to ascertain the level of alcohol involvement in recreational vehicle fatalities in New Zealand, but given that these are rare events, and that apart from the bicycle crashes many occur on farms and are most likely related to farming activity, it seems quite likely that alcohol is not a major contributing factor in these crash events. Introduction This study was part of an international effort to examine existing data sources to determine the role of alcohol in fatalities involving a recreational vehicle, irrespective of the nature of the activity engaged in at the time of the crash

4 The use of certain types of recreational vehicles is common in New Zealand. For example, since their introduction in the early 1970s, ATVs (mainly 4- wheel motorcycles) have become very popular with approximately 7,000 sold each year and an estimated 70,000 currently in use (1). ATVs are extensively used in agriculture, and given that there are an estimated 66,000 farm units in New Zealand, (2) that equates to approximately one ATV per farm unit, or alternatively one ATV for every 50 people in the total population. On farms, ATVs are often referred to as farm bikes, which is a generic term used for any type of motorcycle used on a farm (3). Over the past few decades the most common type of vehicle used as a farm-bike has changed. In the 1970s it was typically a two-wheeled motorcycle purchased mainly, if not solely, for farm use. In the 1980s the two-wheeled farm bike began to be replaced by the three-wheeled version, which because of its inherent instability was soon replaced by the now ubiquitous four-wheeled farmbike. In addition to their use in the farming sector, ATVs are also used in the forestry industry and the tourism sector, and to a lesser extent as private recreational vehicles (1). Bicycles are also widely used and in 2000, there was an estimated 750,000 bicycles in New Zealand (4). In 1996 it was estimated that New Zealanders cycled 351 million kilometres, with year olds accounting for nearly 30% of this travel (4). Most cycling is for recreation or travel to school, with only 3.1% of New Zealanders travelling to work on a bicycle Other types of recreational vehicles are not common in New Zealand. Unlike in Canada (5), snowmobiles are not widely used in New Zealand and few other land-based recreational vehicles are commonly used. Previous research in New Zealand, relevant to the topic of recreational vehicle injury, has examined injuries associated with motorcycle use on road (6) motorcycle and ATVs use off-road (3) and bicycling (7). To date, however, there has been no attempt to provide an overall picture of recreational vehicle fatalities. The role of alcohol as a contributing factor in fatalities, other than motor vehicle traffic crash fatalities, has received scant research attention in New Zealand. One exception is a recently published study of drowning and alcohol, where the coronial files were examined to determine the quality and completeness of alcohol data available in these files (8). The authors reported that the level of information on the coronial files was insufficient to arrive at an accurate estimate of the role of alcohol in drownings, although they considered that is may not apply to other causes of death. In another study of work-related fatal injuries in the agricultural industry, (2) the recording of alcohol involvement in the coronial files indicated alcohol or other drug involvement in 5.6% of the work-related farm injury fatalities (personal communication, Simon Horsburgh). Many of the records did not include any mention of alcohol, therefore this would have to be considered a conservative estimate. The aim of the present study was to examine existing electronic databases to (1) identify fatal crashes involving a recreational vehicle (as defined for this study), irrespective of whether it was being used for recreation, and (2) to determine the extent to which the involvement of alcohol could be ascertained from these data. This was an exploratory study, and relied on readily accessible existing databases. Unlike in Canada, the official police traffic crash reports in New Zealand do not include off-road crashes,

5 and therefore many crashes involving recreational vehicles are not included in the official traffic crash records. Furthermore, for traffic crashes, the vehicle type recorded in the crash reports does not permit identification of recreational vehicles. Also, the traffic crash reports only include bicycle crashes if a motor vehicle is involved, so these too will be undercounted in the official records. Another potential source of off-road crash fatality data was the New Zealand Health Information Services (NZHIS) mortality databases. This file contains records for all fatalities that occur in New Zealand, and these are E-coded which allows identification of off-road crashes. In addition to the E-codes, each record on this file includes a free text descriptor which provides additional information about the injury incident. Although the type of information recorded is not standardised, it can help with identifying the type of vehicle, the activity engaged in at the time of the crash, and alcohol involvement. The coronial data, is another source of data that has been used in some previous studies in New Zealand (2, 8), but this is not recorded electronically so was not readily available and accessible for the purposes of the present exploratory study. Method The data used to identify recreational vehicle fatalities in New Zealand was the New Zealand Health Information Services (NZHIS) mortality database. The NZHIS maintain a record of all injury fatalities that occur within New Zealand. These records are given an external cause of injury code (E-code) (9) which permits the identification of traffic (E ) and non-traffic crash (E ) fatalities. The E-code also includes a fourth digit that indicates that road user status of the victim. The following fourth digit classifications were used in the present study to identify recreational vehicle fatalities for E-codes : 0 driver of a motor vehicle other than a motorcycle; 1 passenger of a motor vehicle other than a motorcycle; 2 motorcyclist (driver); 3 pillion passenger; 6 pedal cyclist. For E-codes the fourth digit of 1 was used to identify bicyclist. For the present study the following NZHIS E-coded data was used: 1. E : all off-road motor vehicle fatalities 2. E826: pedal cyclist, non-motor vehicle, on road and off road. 3. E on-road pedal cyclist versus motor vehicle 4. E free text was examined to identify on road ATV drivers 5. E829 other road vehicle crashes The NZHIS file does not include an indicator of alcohol involvement, but does include a narrative description of the injury event. This is an unstructured free text of up to 90 characters, and provides information about the circumstances of the injury event, and may include information on alcohol. Blood alcohol concentrations can be recorded using a $ sign, for example 130mg/100ml of blood would be recorded as $130, or it may be recorded as text, or both. For this study, the free text was manually examined for any indication of alcohol. Results For the five years , 115 recreational vehicle fatalities were identified from the NZHIS file, with fatalities occurring each year. Table 1 provides the E-code and a description of the types of crashes involved and shows that 69% (n=80) of the fatalities were bicyclists. For all age groups, particularly among the younger age groups, bicyclist crashes on road (E813.6) were

6 the most common type of crash. They accounted for 25 (76%) of the fatalities in the 0-14 age group, 15 (63%) of the fatalities in the15-24 year age group. The next most common category was motorcyclist driver off-road, followed by motor vehicle driver (not motorcycle) off-road. Some inconsistency was noted in the coding for the crashes where the driver was driving a 4 wheel motorcycle with it being coded as either (motor vehicle driver (not motorcycle) off-road ) or as (motorcyclist driver off-road). Excluding the bicycle crashes, a 4-wheeled motorbike was involved in 21 (60%) of the fatalities, a 3-wheeled motorbike, a farm-bike, a trail bike and a motorcycle. Table 1: Description of recreational vehicle fatalities, by E-code, E-code Description N % Motorcyclist (driver) on-road Bicyclist collision with motor vehicle Motor vehicle driver (not motorcycle) off-road Motor vehicle passenger (not motorcycle) off-road Motorcyclist (driver) off-road Unspecified person off road Bicyclist no collision with motor vehicle 10 8 Total Table 2: Age and gender of fatalities, Age (in years) Number % Total Gender Male Female Total (number of wheels not specified) were involved in one fatality each. The other vehicles involved included a beach-buggy, a golf cart, stock cars, and a drag racing car. There were no fatalities identified for E-code 829 that met the criteria for this study. From the event description, the place of occurrence for the off-road crashes was: farm (n=13), race-track (n=6), beach (n=1), and over bank or over hill, not further specified (n=3). The age and gender of the fatalities are given in Table 2 and shows that they were predominantly male, and 50% were aged 24 years or less

7 Of the 115 cases, alcohol involvement was recorded for only 6 cases (5%), with 5 recording a positive BAC. One of the positive BACs was a 14 year old BMX bicyclist with a reading of 70mg/100 ml. Discussion For each year of the five years included in this study, the rate per 100,000 population for fatalities related to recreational vehicles in New Zealand was approximately 0.6. Over the same period the motor vehicle traffic crash fatality rate ranged from 13.9 in 1994 to 11.5 in This indicates that recreational vehicle fatalities are comparatively rare events in New Zealand. Many of the ATV fatalities occurred on a farm, which indicates that these vehicles were most likely being used for farm work-related activity. This is the common use for these types of vehicles in New Zealand. Many farm activities that traditionally required a tractor are now undertaken using a 4-wheel motorcycle, often with a trailer attached. Given the relatively hilly nature of the New Zealand countryside, it was not unexpected, therefore, that the majority of the ATV fatalities occurred on a farm. The extent to which alcohol was involved in the recreational vehicle fatalities could not be reliably estimated due to the limited data contained on the NZHIS file. Coronial files may be a better source of these data, although the experiences of Warner et al. (8) suggest that this too may not provide the quality and completeness of data required to make an accurate estimate of alcohol involvement to be made. In New Zealand a post-mortem BAC is not consistently obtained, and there are few guidelines as to who should be tested. Warner et al. (2000) recommended that all drowning victims 10 years and older should have BAC test done. The results from the present study indicate that this should also include all vehicle related fatalities, and possibly all injury related deaths, irrespective of where they occur. Also, the results from the coronial BAC test should be recorded on the NZHIS file, irrespective of the results of the test. This would greatly facilitate studies such as the present one, and provide reliable estimates of the role of alcohol in this important area of public health. Acknowledgments The assistance of Nicola Dow, data manger at the Injury Prevention Research Unit is acknowledged, and also the New Zealand Health Information Service for the use of the data. References 1. J. Wren. All terrain vehicles (ATV) Factsheet. Safekids, Auckland, New Zealand, S. Horsburgh, A.-M. Feyer, J. D. Langley. Fatal work related injuries in agricultural production and services to agriculture sectors of New Zealand, Occupational & Environmental Medicine 2001; 58: J. D. Langley, S. W. Marshall, D. J. Begg, A. I. Reeder. Motorcycle & ATV crashes in New Zealand resulting in death and hospitalization: non-traffic crashes. Journal of Agricultural Safety and Health 1995; 1:

8 4. Land Transport Safety Authority. Cycle safety in the spotlight for national bike week, in Road Safety. Land Transport Safety Authority, Wellington, D. J. Beirness, K. Desmond, Alcohol involvement in recreational vehicle fatalities in Canada. T International conference on alcohol, drugs, and traffic safety, Stockholm D. J. Begg, J. D. Langley, A. I. Reeder. Motorcycle crashes in New Zealand resulting in death and hospitalisation I: introduction, methods, and overview. Accident, Analysis and Prevention 1994; 26: B. A. Collins, J. D. Langley, S. W. Marshall. Injuries to pedal cyclists resulting in death and hospitalisation. New Zealand Medical Journal 1995; 106: M. Warner, G. S. Smith, J. D. Langley. Drowning and alcohol in New Zealand: what do the coroner's files tell us? Australian and New Zealand Journal of Public Health 2000; 24: World Health Organisation, International classification of diseases. World Health Organisation,

9 Ontario Provincial Police Specialized Patrols Trails and Waterways Enforcement Snowmobile All-Terrain Vehicle Vessel Enforcement (S.A.V.E.) Teams 1 J.H. McDonnell, 2 C. Hughes and 3 L. Lee-Davidson 1 Ontario Provincial Police, North East Regional Headquarters, North Bay, Ontario 2, 3 Ontario Provincial Police, General Headquarters, Orillia, Ontario Keywords Recreational vehicles, alcohol, collisions, police, countermeasures Abstract This report will provide the background with regard to the problem of alcohol abuse in Ontario by some recreational vehicle users. Recreational vehicles include snowmobiles, vessels, and All- Terrain Vehicles (ATVs). The article will also illustrate the typical profile of individuals that engage in the abuse of alcohol while operating recreational vehicles. To address this serious problem, the Ontario Provincial Police received additional provincial funding in 2001 to form the Snowmobile ATV Vessel Enforcement (S.A.V.E.) Unit. The three SAVE teams are staffed with experienced OPP law enforcement officers and are located strategically in the Province. The specialized unit supplements the efforts of regular front-line police officers and focuses on the reduction and prevention of recreational fatalities through enforcement and educational presentations. Introduction By virtue of Ontario s immense size, generous resources, diverse terrain, and thousands of lakes, the Province is an ideal year round recreational playground. There are almost 4000 lakes in Ontario that are larger than 3 km 2 and over km of snowmobile trails that are maintained by the Ontario Federation of Snowmobile Clubs. There is also an equal distance of local (unofficial) trails throughout the Province. Ontario supports a significant resource based tourism industry, pumping billions of dollars into the provincial economy. Recreational vehicle operators contribute a significant percentage to this industry through boating, fishing, hunting, and traveling on a network of trails throughout the province. According to a 1996 Canadian Coast Guard study 1, an estimated 1.4 million people either owned or rented boats in Ontario. As well, Ontario is a popular destination for recreational boaters from other provinces and also from the United States. The neighbouring eight Great Lakes states account for an estimated additional 3.8 million registered recreational boats. In 1999, there were also approximately Motorized Snow Vehicles (MSV) and Off-Road vehicles

10 registered in Ontario (ORSAR). 2 outselling MSVs. Manufacturer s data indicates that ATVs are significantly The Ontario Provincial Police (OPP) is mandated by provincial legislation to provide policing services to over 95% of the Province s waterways. The OPP is also the primary search and rescue agency for most of those areas. Similarly, the majority of Ontario s snowmobiling activities and trail systems are found within OPP jurisdiction. The OPP also provides policing services to most of the smaller communities in rural and northern Ontario communities where the use of a recreational vehicle is a way of life. Operators of recreational vehicles from larger urban centers are more likely to travel to these areas on either a seasonal or a weekend basis. Historically, recreational riders in some areas have become accustomed to limited police patrols and enforcement on the trails and waterways. The chance of encountering law enforcement officials was virtually non-existent in those areas of the province once the operator left the shoreline or embarked on a trail. Not only was this a factor for drinking/riding, but the lack of enforcement indirectly supported a norm in which alcohol consumption and the operation of a recreational vehicle became part of the same activity. In other words, a culture of alcohol abuse was accepted and tolerated by those riders. The consequences of such abuse have resulted in tremendous carnage on the trails and waterways. Risk-taking individuals prone to alcohol abuse often placed themselves and other innocent people in jeopardy. The challenge for law enforcement officials is to alter the unacceptable and unlawful behaviour that has become so prevalent. The number of fatalities resulting from recreational vehicle use varies from year to year. Environmental conditions are a major factor in the determination of the length of the season. Injuries resulting from crashes are not known because many incidents are not reported to the police. Bierness cites Rowe et al 3 in suggesting that the rate of snowmobiling injuries may be as high as 17 times greater than official estimates. The use of alcohol while operating a recreational vehicle is well established as being one of the main factors in boating and snowmobiling incidents. During the 2000/2001snowmobile season, 36 lives were lost in snowmobiling fatalities in OPP jurisdiction. Despite several years of reduction in alcohol-related deaths, alcohol remained a factor in 40% of those incidents. In the year 2001 boating season, alcohol was a factor in 50% of the marine fatalities investigated by the OPP. Finally, alcohol was identified as a contributing factor in 16.7% of the all-terrain vehicle deaths that occurred in OPP jurisdiction in All of these incidents have occurred despite the existence of a Criminal Code offense of operating of any type of motorized recreational vehicle or vessel while impaired or with a BAC exceeding.08. By contrast, the percentage of fatal road crashes investigated by the OPP in which alcohol was a factor is significantly lower. 5 In order to illustrate the magnitude of the problem of alcohol abuse in relation to the operation of recreational vehicles in Ontario, the following six tables have been produced. Table 1 depicts the relationship of alcohol as a contributing factor in all motorized vehicle/marine vessel deaths in Ontario (OPP jurisdiction) in Table 2 depicts the 5 year trend in Ontario deaths relating to the operation of snowmobiles from the current 2001/02 season back to the 1997/98 season. Table 3 depicts deaths relating to marine incidents in Ontario from 1997 to Table 4 depicts All

11 Terrain Vehicle (ATV) deaths in Ontario from 1997 to Table 5 depicts the age/gender relationship of fatalities in all motor vehicle deaths in 2000/2001. Table 6 is a statistical record of the SAVE Unit s efforts in several categories. As indicated in the following tables, each year an average of 46 persons are killed in boating mishaps, 25 people on snowmobiles in OPP jurisdiction and 8 persons on ATV s. Causal factors remain constant alcohol consumption, poor operating practices, and/or inappropriate use of mandated safety equipment (e.g. lifejackets, helmets). Table 1: Relationship of alcohol as a contributing factor in fatalities in 2001 (OPP Jurisdiction) Marine All-Terrain Vehicle SnowVehicle Motor Vehicles Alcohol 42.9% 16.7% 4% 19.6% Speed 5% 22.9% 24.9% Other 4% 33.3% 37.1% 55.5% Unknown 17.1% Table 2: 5-year trend of snowmobile fatalities, 2001/02 season to 1997/98 season (OPP Jurisdiction) 2001/ / / / /98 5-yr Average Number of fatalities % Alcohol-related % on OFSC trails Table 3: 5-year trend of waterway fatalities, 1997 to 2001 (OPP Jurisdiction) yr Average Number of fatalities % Alcohol-related % not wearing PFD Table 4: 5-year trend of all-terrain vehicle fatalities, 1997 to 2001 (OPP Jurisdiction) yr Average Number of fatalities % Alcohol-related % not wearing helmet

12 Table 5: Percentage distribution by age/gender of fatally injured persons in recreational vehicles, 2001 by comparison with distribution of roadway fatalities in OPP jurisdiction. (yrs of age) Marine All-Terrain Vehicle Snowmobile Motor Vehicle Male Under Over Female Under Over It appears that there is a high-risk group of recreational vehicle operators who share similar characteristics (age/gender) and engage in similar types of behaviour (alcohol, speed, and lack of protective equipment). Through statistical analyses of fatal incidents, the high-risk operator profile that has emerged is a male between 15 and 49 years of age, consuming alcohol, riding on weekends alone or with friends, on lakes and rivers at night and driving too fast for the conditions. Similar characteristics are also identified in many high-risk drivers on the highways in Ontario. Several provincial and national programs have been developed to combat the high-risk behaviours in operating recreational vehicles. These programs combine elements of education, enforcement, engineering, and media coverage. Some of these specific programs include the Snowmobile Trail Officer Patrol (STOP), Sled Smart, and the North American Safe Boating Campaign. In addition, many local initiatives have been developed to address specific concerns or issues within various communities. Traditionally, other competing pressures for policing services have created significant challenges for police managers to deal with the issues outlined in this report. For example, in the past, there has been limited marine policing by the OPP in the spring and fall despite the many fatalities that occur during those periods. Enforcement is dependent upon the availability of law enforcement personnel. Competing priorities for front-line service combined with policy requiring two OPP staff in a boat make it difficult for police managers to operationalize marine patrols during these non-peak times. The formation of the Ontario Provincial Police SAVE Unit is a strategy that provides policing resources through specialty a unit to address these issues on a full-time basis throughout the entire year

13 Methods The Ontario Provincial Police SAVE Unit is comprised of eighteen members. Each of the three SAVE teams is staffed with five Provincial Constables and a Sergeant supervisor. The three SAVE teams are strategically located in Orillia, North Bay, and Kingston. A Provincial SAVE Coordinator monitors operational assignments and initiatives to ensure that resources are utilized in a manner that is equitable and efficient. Members of the SAVE unit are fully equipped and trained to respond to marine, snowmobile, and ATV incidents. Each member of the unit is provided with personal equipment that includes snowmobiles, ATVs, and vessels. The SAVE Unit is tasked with a provincial mandate and also serves as a provincial resource to the 85 OPP detachments across the Province. A SAVE team may be requested by detachment personnel or community groups to supplement local programs and to support new initiatives. SAVE unit members also provide training to other OPP members on specialized topics and equipment. SAVE members are expected to participate in educational seminars with marine, snowmobile, and ATV stakeholders during the year. They are also expected to create public awareness of the teams and the key recreational vehicle safety issues through visibility at boat, cottage and industry shows. Through their participation at large events, SAVE team members have the potential to interact with over persons each year. They also actively participate in local initiatives, augmenting various initiatives in the many communities policed by the OPP throughout Ontario. SAVE teams can also fill in gaps in current programs by providing resources for educational or informational purposes. The three teams also are expected to maintain a public profile through attendance at winter carnivals, Ontario Federation of Snowmobile Club functions and other special events. Finally, the teams also engage in their own initiatives as a six-person team or an eighteen-member unit to target problem areas within OPP jurisdiction. Results The three OPP SAVE teams have been operational since June 11, Since that time they have performed preventative patrol and proactive educational presentations across the Province. The following table illustrates the levels of enforcement and numbers of presentations that have resulted from the unit s efforts. Table 6: 2001 total SAVE Unit combined enforcement numbers (Marine, ATV and Snowmobile Patrol). Patrol Hours Public Service Hours Presentations/Public Events Vessels/Vehicles Checked Number of Charges Number of Warnings 2001 (June Dec.) 2, ,136 1,140 1,904 Although the data in Table 6 is impressive, the outcomes that are directly attributable to the SAVE Unit s efforts will be difficult to measure in the short term. The impacts will need to be measured over time, and factors such as environmental conditions, advancements in safety

14 equipment, community involvement, and potential impacts of any new legislation will also have to be considered and analyzed. In the meantime, evidence as to the impact of the SAVE unit is based on anecdotes and single incidents that suggest the culture of recreational vehicle riding is being gradually changed. For example, in many communities, it has been traditionally accepted to hold events in which amateur snowmobile riders perform daring crossings of open water for crowds of onlookers. In the winter of 2002, the SAVE teams challenged this practice from a safety/legal viewpoint. Through the catalyst of the SAVE involvement and the cooperation of judicial officials and municipal leaders, efforts are being coordinated to limit or stop this dangerous practice. This example illustrates how the standard of behaviour is being altered through the intervention of the SAVE teams to prevent serious injuries or deaths that may result. Discussion The SAVE Unit has been in existence in Ontario for just over a year. Although it is premature to analyze the effect on fatal crashes and injury collisions involving recreational vehicles from a statistical standpoint, the Unit anecdotally has had a dramatic effect in some of the problem areas in the Province. In the future, the SAVE Unit will continue to develop other program strategies that target core safety issues. The SAVE Unit will also provide consistency in the awareness, enforcement, and education surrounding issues involving the operation of recreational vehicles in Ontario. References 1. Lifesaving Society. The Ontario Drowning Report, 2000 Edition. Page Road User Safety Branch. Ontario Road Safety Annual Report: Ontario Ministry of Transportation. ISSN# Beirness DJ. Alcohol Involvement in Snowmobile Operator Fatalities in Canada. Canadian Journal of Public Health; 92: Hughes, C. Traffic & Marine Trend Analysis: Ontario Provincial Police, Field & Traffic Support Bureau. 5. Hughes, C. Traffic & Marine Trend Analysis: Ontario Provincial Police, Field & Traffic Support Bureau

15 Drinking and Boating: a Population-Based Case- Control Study of Recreational Boating Fatalities Gordon S. Smith, M.D,MPH, Penelope M. Keyl, M.Sc., Ph.D., Jeffrey A Hadley, Ph.D. Christopher L. Bartley, M.A., Robert D. Foss, Ph.D., William G Tolbert, M.A, James McKnight, Ph.D. Johns Hopkins Center for Injury Research and Policy, Baltimore, MD (Drs. Smith, Keyl, and Hadley), Johns Hopkins School of Medicine, Department of Emergency Medicine (Drs. Keyl and Smith), Highway Safety Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC (Drs. Foss, Mr. Bartley, and Mr. Tolbert now at Rho Inc. Chapel Hill, NC.), Pacific Institute for Research and Evaluation (PIRE), Rockville, MD (Dr. McKnight) Abstract Background: Whereas alcohol is involved in many boating fatalities, the relative risk of death associated with alcohol use among boaters is unknown. Objectives To determine the association of alcohol use with the relative risk of dying while boating for passengers and operators. Methods A case-control study was conducted where recreational boating deaths (18 years and older) during the period from 1990 to 1998 in Maryland and North Carolina (N=221) were compared with controls obtained from a multi-stage probability sample of boaters in each state during (N=3943). Logistic regression was used to calculate relative risk of a boating fatality associated with different levels of blood alcohol concentration (BAC) while controlling for other risk factors. Stratification, clustering, and weighting were also taken into account in the analysis. Results Compared to a zero BAC, the relative risk of death was increased even at 10 mg/dl, with an estimated OR of 1.3 (95% CI: ); increasing to 52.4 (95% CI: ) at 250 mg/dl; and was similar for passengers and operators. The relative risk associated with alcohol use did not vary by boat type and was the same whether the boat was moving or was not underway

16 Discussion Drinking increases the relative risk of dying while boating even at low levels of BAC. In addition, drinking by passengers as well as operators is associated with the same increased relative risk of death regardless of whether the boat is underway. Prevention efforts targeted only at those operating a boat are missing many boaters at high risk. These findings suggest that countermeasures directed only at operators of moving boats are likely to have less impact on alcohol-related boating fatalities than broader efforts to address drinking by anyone engaged in recreational boating

17 Alcohol Involvement in Recreational Vehicle Operator Fatalities in Canada D.J. Beirness Traffic Injury Research Foundation of Canada Abstract This paper provides an update on previous work that examined alcohol use among fatally injured operators of three types of recreational vehicles snowmobiles, bicycles and all terrain vehicles (ATVs). It uses data from the TIRF Fatality Database from 1987 through 1999 to compare alcohol involvement in fatally injured operators of these vehicles to that among fatally injured drivers of automobiles. The implications of the findings for programs and policies are discussed. Introduction The use of recreational vehicles appears to have become increasingly popular over the past decade. For example, the Canadian Council of Snowmobile Organizations (1) estimates that in 1999 there 760,000 registered snowmobiles in Canada. Snowmobile enthusiasts have formed 914 snowmobile clubs representing about 500,000 family snowmobilers across the country. These clubs operate and maintain an expanding network of over 132,000 kilometers of organized snowmobile trails. The Canadian Cycling Association reports that a recent poll found that 66% of Canadians age 18 and over (about 13.7 million people) are non-competitive cyclists (2). They also report that increased bicycle sales and participation in cycling events suggest that there are more cyclists now than ever before in Canada. The fastest growing segment of the cycling community are teens and adults. Casual observation indicates an increased popularity of all-terrain vehicles. The all-season versatility of these vehicles makes them particularly attractive as a form of recreation. No information was available on the extent of use of these vehicles. Although these vehicles are referred to in this paper as recreational vehicles, it should be noted that they are often used for other purposes. For example, ATVs are used in farming and transportation in areas not easily accessible to conventional vehicles. Snowmobiles serve a similar function in winter and are also a form of primary transportation in some areas. Bicycles are often used for commuting as well as recreation. The growth in recreational vehicle use has been associated with increased concern about deaths and injuries as a result of crashes. The actual number of persons killed and injured in crashes involving a recreation vehicle can be difficult to ascertain. Although motor vehicle departments routinely capture data on crashes involving these vehicles that occur on public highways, similar data are not routinely available on those crashes that occur on private property. Hence, little is known of the factors that contribute to recreational vehicle crashes, including the role of alcohol

18 The use of alcohol by recreational vehicle operators has become an issue of increasing concern. Alcohol is commonly consumed in association with recreational activities. When the activity happens to involve the operation of a vehicle, the combination can have particularly tragic consequences. In recognition of the inherent dangers of operating a motor vehicle under the influence of alcohol, the Criminal Code of Canada indicates that it is an offence to operate any type of motor vehicle while impaired or with a BAC in excess of 80 mg%. It does not matter whether the vehicle is being operated on a public roadway or on private property. The law and its penalties are applied to snowmobile and ATV operators the same as they are to drivers of highway vehicles. The one notable exception is that operators of bicycles have been specifically excluded from the legislation. Over the past several decades, research has clearly established that alcohol can and does impair the ability to operate a motor vehicle. Alcohol involvement in highway crashes has also been well-documented. Considerably less attention has been paid to the role of alcohol in fatal crashes involving other types of vehicles, particularly those that are often used for recreational purposes and not necessarily on public roadways. The impairment of the ability to operate a vehicle undoubtedly applies to all types of vehicles, not just those that travel public roadways. This paper examines the incidence of, and trends in, alcohol use among fatally injured operators of three types of recreational vehicles snowmobiles, bicycles, ATVs in Canada over the 13- year period from 1987 through 1999 and compares alcohol involvement among fatally injured recreational vehicle operators to that among fatally injured drivers of automobiles. Method The primary data on vehicle operator fatalities were obtained from the Fatality Database of the Traffic Injury Research Foundation (3). These data, collected from all provinces and territories in Canada, include information from coroners files on the results of toxicological tests for the presence and amount of alcohol among persons fatally injured in motor vehicle crashes. A unique feature of this database is that it includes not only those fatalities that occur on public roadways but those that occur on private property as well. This paper reports data on recreational vehicle operator fatalities (including snowmobiles, bicycles, and ATVs) age 16 years and over for the years 1987 through Fatally injured drivers of automobiles were used as a comparison. Results Over the thirteen-year period from 1987 to 1999, a total of 2,998 persons died while riding a bicycle, snowmobile or ATV. During this same period, 26, 883 persons died in automobiles. To examine the involvement of alcohol in these fatalities, cases were restricted to fatally-injured operators of these vehicles who were at least 16 years of age and who were tested for alcohol. Testing rates for alcohol were 79.2% for automobile drivers; 71.1% for bicyclists; 81.5% for snowmobile operators; and, 69.7% for ATV drivers. Only those cases tested for alcohol were included in the following analyses. The number (and percent) of fatally injured drivers of each type of vehicle who tested negative for alcohol and the number (and percent) who tested positive for alcohol are presented in the first two columns of Table 1. Overall, snowmobile operators had the highest incidence of alcohol involvement at 71.2%. This was followed by operators of ATVs at 55.7%. Among fatally

19 injured bicyclists, 24.7% were found to have been drinking. In comparison, alcohol was found in 43.2% fatally injured automobile drivers. In Table 1, the column labeled Odds indicates the odds of a fatally injured driver of that type of vehicle testing positive for alcohol. For example, a fatally injured snowmobile operator is 2.47 times more likely to have been drinking than not. Similarly, fatally injured ATV drivers are 1.26 times more likely to have been drinking than not. On the other hand, the values of the odds ratios for automobile drivers and bicyclists are both less than one, indicating that these drivers are less likely to have been drinking than not drinking. Table 1: Number, percent, and relative odds ratios of fatally injured drivers of four vehicle types according to alcohol use BAC=0 Total Odd Relativ Confidence BAC>0 s e N (%) N (%) N Odds Intervals CARS (56.8) (43.2) 4 BICYCLES 363 (75.3) 119 (24.7) SNOWMOBILE 214 (28.8) 529 (71.2) S ATV 113 (44.3) 142 (55.7) The relative odds provide a means to compare the likelihood of alcohol involvement among fatally injured operators of recreational vehicles to that among fatally injured automobile drivers. The relative odds is simply the ratio of the odds of alcohol involvement in driver fatalities of one type of vehicle to the odds of alcohol involvement in automobile driver deaths. The 95% confidence interval provides an indication of the statistical significance of the relative odds. These findings indicate that fatally injured snowmobile operators are 3.26 times more likely to have been drinking than fatally injured car drivers. The likelihood that a fatally injured driver of an ATV had been drinking is 1.66 times more likely than a car driver to have consumed alcohol. On the other hand, fatally injured bicyclists are less likely to have been drinking than car drivers. The distribution of BAC among fatally injured operators of each of the four types of vehicles is displayed in Table 2. At each level of BAC, the percent of fatally injured snowmobile and ATV operators is greater than that of fatally injured automobile drivers. The reverse is true for bicycle riders. Their BAC levels are typically lower than those among automobile drivers. Within each type of recreational vehicle, however, the distribution of BAC reveals a pattern similar to that among automobile driver fatalities. The proportion of drivers decreases from the 1 to 49 mg% range to their lowest values in the 50 to 80 mg% range, then increases in the 81 to 150 mg% range. Drivers with BACs in excess of 150 mg% comprise the largest category of alcoholpositive fatalities

20 Table 2: Number (and percent) of fatally injured operators of four vehicle types according to BAC Total BAC=0 BAC=1-49 BAC=50-80 BAC= BAC>150 N (%) N (%) N (%) N (%) N (%) CARS (56.8) 666 (4.9) 357 (2.6) 1224 (9.0) 3619 (26.6) BICYCLES (75.3) 14 (2.9) 8 (1.6) 23 (4.8) 74 (15.3) SNOWMOBILES (28.8) 72 (1) 36 (4.8) 133 (17.9) 288 (38.8) ATV (44.3) 20 (7.8) 8 (3.1) 34 (13.3) 80 (31.4) Trends in the percent of alcohol-involved operator fatalities are presented for each type of vehicle in Figure 1. Alcohol-involvement among fatally injured automobile drivers has declined progressively over the period examined from 50% in 1987 to 30% in This trend continues that which was so prominent during the 1980s (4). Among fatally-injured bicyclists and ATV operators, alcohol involvement fluctuates considerably from year to year, largely due to the relatively small number of cases. Overall, there has been no substantial changed since Alcohol involvement in snowmobile operator fatalities has decreased from over 80 % in the late 1980s to just under 70% in more recent years. Figure 1: Percent of Fatally Injured Operators Positive for Alcohol Percent Bicycles Snowmobiles ATVs Automobiles The number (and percent) of fatally injured drivers of each type of vehicle who tested negative for alcohol as well as the number (and percent) who tested positive for alcohol in five age groups are displayed in the first two columns of Table 3. Overall, alcohol involvement was highest among fatally injured drivers aged 21 to 25 and 26 to 35 across all vehicle types. Drivers over 55 years of age had the lowest percent of alcohol involvement

21 The relative odds and 95% confidence intervals are presented in the last two columns of Table 2. The relative odds provide a means to compare the likelihood of alcohol involvement in fatal crashes of recreational vehicles to that in automobile crashes within each age group. These findings indicate that in every age group, fatally injured snowmobile operators are significantly more likely to have been drinking than fatally injured car drivers. The likelihood that fatally injured drivers of ATVs had been drinking is significantly different from that of car drivers only for those aged 26 to 35 and 36 to 55. Bicycle riders were less likely than car drivers to have been drinking in all age groups. Table 3: Number, percent, and relative odds ratios of fatally injured drivers of four vehicle types according to alcohol use and age Age Group BAC=0 BAC>0 Total Odds Relative Confidence N (%) N (%) N Odds Intervals CARS 1126 (55.0) 920 (45.0) BICYCLES 691 (82.1) 15 (17.9) SNOWMOBILES 31 (36.0) 55 (64.0) ATV 35 (64.8) 19 (35.2) CARS 899 (42.6) 1211 (57.4) BICYCLES 48 (68.6) 22 (31.4) SNOWMOBILES 32 (22.9) 108 (77.1) ATV 15 (38.5) 24 (61.5) CARS 1377 (45.1) 1679 (54.9) BICYCLES 93 (73.2) 34 (26.8) SNOWMOBILES 50 (18.9) 215(81.1) ATV 15 (26.8) 41 (73.2) CARS 2064 (58.0) 1495 (42.0) BICYCLES 77 (68.1) 36 (31.9) SNOWMOBILES 66 (33.8) 129 (66.2) ATV 20 (31.3) 44 (68.8) >55 CARS 2252 (80.2) 556 (19.8) BICYCLES 73 (89.0) 9 (11.0) SNOWMOBILES 34 (64.2) 19 (35.8) ATV 27(67.5) 13 (32.5) Discussion Over the past two decades, the unprecedented attention directed towards the alcohol-crash problem focused almost exclusively on drivers of passenger vehicles on public highways. The data presented in the present paper highlight the fact that the problem does not end where the road stops. In fact, the proportion of alcohol-related snowmobile and ATV operator fatalities is

22 higher than that among fatally injured drivers of automobiles in Canada. Although some of these fatalities occurred as a result of a collision on a public highway, the majority occurred in off-road locations. Bicyclist fatalities present a different picture. Fatally injured bicyclists are significantly less likely than automobile driver fatalities to involve alcohol. The relatively low rate of testing for alcohol among bicyclist fatalities, however, warrants caution be used in the interpretation of this finding. Nevertheless, the fact that approximately one-quarter of adult bicyclist fatalities involve alcohol use by the rider, and that most of these involve high BACs, is cause for concern. Bicyclists are vulnerable vehicle users, particularly when they interact with road traffic. The use of alcohol would undoubtedly impair the balance and coordination necessary to control the vehicle, thereby compounding the risk of death and injury. Decades of research in the area of alcohol and traffic safety has clearly established that drinking drivers are at significantly greater risk of being involved in a serious motor vehicle crash than drivers who have not consumed alcohol. At this point, however, it is only an assumption that the same relationship holds with regards to recreational vehicle operators. The type of case-control studies that established the link between alcohol and highway vehicle crashes have not been done with recreational vehicles. Although it is probably safe to assume that alcohol, consumed in sufficient quantities, can impair the ability to operate a recreational vehicle safely, the degree of risk has not been established. The data presented in this paper provide evidence that alcohol is present in a relatively high proportion of recreational vehicle operator deaths each year; the extent to which alcohol is present among operators of these vehicles who do not crash is unknown. Without this latter piece of information, it is not possible to determine the extent to which alcohol is over- (or under-) represented among recreational operator fatalities. Random breathtesting surveys of recreational vehicle operators would be one means of helping to fill the information gap. The lack of definitive studies on the risks of drinking and operating a recreational vehicles should not constrain efforts to enhance safety. Specific countermeasure initiatives are needed to address the apparently common practice of consuming alcohol as part of recreational vehicle use. For example, the dangers of operating a recreational vehicle after drinking could be outlined in educational programs and informational materials delivered through local snowmobile clubs, bicyclist organizations, manufacturers, suppliers, and retailers at the community level. This may be the most efficient and effective means of targeting awareness and generating concern about drinking and driving among recreational vehicle operators There is also a role for increased enforcement of drinking-driving laws among recreational vehicle operators. Unfortunately, these activities often occur in relatively isolated areas which renders typical enforcement efforts difficult and inefficient. Greater use of enforcement teams travelling on similar types of vehicles may be an alternative worthy of consideration. Acknowledgement The TIRF Fatality Database is supported by the Canadian Council of Motor Transport Administrators (CCMTA) and Transport Canada