Characteristics of Older Motorcyclist Crashes

Abstract

In the U.S. as well as other countries, the number of motorcyclists killed in traffic crashes has risen sharply over the past five years, due in part to the increased popularity of motorcycling among older riders. This paper examines trends in motorcyclist casualties and vehicle registrations from 1990–2002, based on national and state (North Carolina) motor vehicle crash and vehicle registration data. The data show similar patterns of increased fatalities that parallel a growth in motorcycle registrations. Whereas the number of motorcyclists ages 16–24 declined over the 13–year study period, the number of riders ages 35 and older increased. Three years of recent (2000–2002) NC data are examined to identify salient characteristics of the crashes of these older riders. Results are discussed with respect to approaches for mitigating the increase in motorcyclist deaths and injuries.

Nationally, the number of motorcyclists killed in traffic crashes peaked in 1980. From 1980 through 1997 motorcyclist deaths declined 59%, from 5,144 to 2,116. However, since 1997 motorcyclist deaths have increased at an average annual rate of over 10%, reaching 3,244 deaths in 2002 (NHTSA, 2004a). Preliminary estimates for 2003 point toward an additional 11% increase (NHTSA, 2004b). During this time, overall U.S. traffic fatalities have remained fairly stable. As a result, motorcyclist deaths as a proportion of overall traffic fatalities increased from 5.0 percent in 1997 to an estimated 8.3% in 2003 (NHTSA, 2004b).

The observed increase in motorcyclist fatalities tracks national motorcycle sales and vehicle registration data. According to FHWA estimates, the number of registered motorcycles peaked in the early 1980s, then declined through the mid 1990s before initiating an upswing in 1998 (Ulmer and Preusser, 2003). Similarly, the Motorcycle Industry Council reports growth in motorcycle sales since 1993, but strong double digit growth since 1997 (Motorcycle Network, 2003).

Other trends may also be affecting the increase in motorcyclist deaths. One is the aging of the motorcycle rider population. According to a 1998 industry survey, the typical motorcycle owner is now a 38 year-old married male making $44,000 a year (MIC, 1998). The increased age of motorcycle owners is reflected in crash and fatality data. Thus, the average age of motorcyclists killed in crashes was 29.3 years in 1990, but 36.5 years just a decade later in 1999 (NHTSA, Feb. 2002).

Along with changes in the rider population have come changes in the vehicles themselves. Traditional motorcycles, once the standard, have now been augmented by Cruiser, Sport, and Touring models (NHTSA and Motorcycle Safety Foundation, 2000). Motorcycles have become larger, heavier, and more powerful. In 1990, less than 22% of fatally injured motorcyclists were riding bikes with engine displacements greater than 1,000 cubic centimeters; in 1999, 34% of bikes fell into this category (Shankar, 2001). A disproportional share of these bikes were being ridden by older motorcyclists.

These trends are not unique to the United States. A recent Organization for European Cooperation and Development (OECD) report noted that while overall traffic deaths had declined, a rapid rise in fatalities among older motorcyclists was a concern in France and Great Britain, as well as the U.S (OECD, 2003). Australia has also witnessed a surge in motorcycle sales and crashes, especially among older riders (ATSB, 2003; Christie and Newland, 2001).

Another potential factor in the increase in motorcyclist fatalities in the U.S. is a decrease in helmet usage. Whereas laws requiring riders of all ages to wear helmets are the norm worldwide (IIHS, 2003), since 1997 six U.S. states (Texas, Arkansas, Kentucky, Louisiana, Florida and Pennsylvania) have repealed or weakened their universal helmet laws. Currently only 19 U.S. states (including North Carolina) plus the District of Columbia have laws in place requiring all motorcyclists to wear helmets. Studies have shown that in states where helmet laws have been repealed or weakened, nearly half of all riders do not wear helmets (Ulmer and Preusser, 2003). Nationwide, observed helmet use dropped from 71% in 2000 to 58% just two years later in 2002 (Glassbrenner, 2002). The decline in helmet use is significant, in that helmets have been shown to reduce the risk of fatal injury in crashes by 39% (Deutermann, 2004).

The purpose of the current paper is to clarify age-related trends in motorcyclist deaths and injuries, using national and state (North Carolina) motor vehicle registration and crash data. The paper then examines the most recent (2000–2002) NC crash data in greater depth to identify salient characteristics of crashes involving older motorcyclists. Drawing from both analyses, suggestions are offered to mitigate the trend of increasing motorcyclist deaths and injuries.

METHODS

National data on motorcycle operator fatalities for the time period 1990–2002 were obtained from Federal Highway Administration (FHWA) published reports, based on Fatality Analysis Reporting System (FARS) data. FARS data represent a census of police-reported motor vehicle crashes resulting in one or more fatalities, and are based on reports by all 50 states plus the District of Columbia and Puerto Rico.

Information on motorcycle registrations over the same time period was also retrieved from published FHWA reports. The number of licensed motorcycle operators was not used as a measure of exposure since this information is not available nationally. Also, many persons operate motorcycles without the required endorsement on their license and, conversely, many others may retain a motorcycle endorsement on their license even though they no longer ride.

The national data were compared to North Carolina crash and motor vehicle registration data obtained from the NC Department of Transportation, Division of Motor Vehicles. North Carolina law requires the reporting of any crash involving one or more motor vehicles and resulting in injury or property damage in excess of $1,000 ($500 prior to January 1, 2000). The computerized data are maintained by the NC DMV and routinely analyzed by staff members at the UNC Highway Safety Research Center.

The study initially compares the available NC and national crash data with respect to motorcycle crash and vehicle registration trends. This is followed by a series of descriptive tabulations for the most recent three years of NC crash data (2000–2002), examining differences in motorcycle operator, roadway, environmental, and crash characteristics by motorcycle driver age (16–24, 25–34, 35–44, 45+). Separate crosstabulations are presented for single vehicle (motorcycle only) and two or more vehicle crashes, and significant age trends identified based on chi-square tests of association.

RESULTS

COMPARISON OF NATIONAL AND NC CRASH AND VEHICLE REGISTRATION DATA

North Carolina motorcycle fatality and vehicle registration data were found to mirror the national data with respect to trends over time. Figure 1 presents motorcycle fatality and vehicle registration trends based on national FARS and FHWA data, while Figure 2 presents corresponding information based on the NC data. For both the national and NC data, the recent increase in motorcyclist registrations has been accompanied by an increase in motorcycle driver fatalities. For the North Carolina data, the relatively small number of fatalities (generally 60–100 annually) contributes to greater year to year variability in the data. However, both the national and NC trend lines show a significant upswing in motorcyclist deaths starting in 1998.

Figure 1.

National motorcycle fatalities and registered motorcycles, 1992–2002 (Source: FHWA).

Figure 2.

NC motorcycle fatalities and registered motorcycles, 1990–2002 (Source: NC DMV)

Information from Figures 1 and 2 is combined in Figure 3, which shows fatality rates per registered motorcycle. Nationally, the fatality rate declined from 1990–1997, but increased over the four year period from 1997 through 2000 before again leveling off. NC fatality rates are higher than the national rates, a fact that likely reflects the generally rural nature of the state. There are no consistent trends in the NC fatality rates over the 13-year time span covered by the data. However, also shown are the overall crash rates, which have declined from a peak of 339 crashes per 10,000 registered motorcycles in 1993 to a low of 209 crashes per 10,000 registered motorcycles in 2002.

Figure 3.

National and NC motorcycle fatality rate per 10,000 registered motorcycles, 1990–2002 (Source: FHWA, NC DMV)

Finally, Figure 4 shows trends with respect to the ages of crash-involved motorcyclists in North Carolina. In 1990, at the beginning of the time period examined, motorcyclists ages 16–24 comprised 43% of crash-involved motorcyclists, while riders ages 45 and older comprised only 11%. By 2002, the youngest motorcyclists had declined to 23% of the total, while the oldest had grown to 26%. Overall, there was a three-fold increase in the number of motorcyclists ages 45+ involved in crashes, and a nearly two-fold increase in the number of motorcyclists ages 35–44 in crashes. Meanwhile, the absolute number of motorcyclists in the youngest age category declined by one-third. Very similar results have been reported elsewhere, based on national FARS data: the only motorcyclists experiencing an increase in fatalities from 1990–2002 were riders ages 40 and above (IIHS, 2003).

Figure 4.

North Carolina motorcycle crashes, by age of motorcyclist, 1990 – 2002

Based on these data, it is clear that programs and policies directed at reducing motorcyclist injuries and deaths must target older as well as younger riders. The following section highlights differences in crashes among the various age groups to provide guidance on the most effective intervention strategies for each.

MOTORCYCLE CRASH CHARACTERISTICS BY OPERATOR AGE

The crosstabulations in this section are based on three years of North Carolina police-reported crash data. During this three-year period, there were 7,532 motorcyclist drivers ages 16 and above involved in crashes. Of these, 48 (0.6%) had missing age information. The remaining 7,484 form the basis for the tables. This number includes 1,856 riders (24.8%) ages 16–24, 2,220 (29.7%) ages 25–34, 1,664 (22.2%) ages 35–44, and 1,744 (23.3%) ages 45+.

Table 1 shows age differences with respect to other descriptive characteristics of the motorcyclist. All of the results shown are statistically significant at p≤.01, based on chi-square tests of association for 2×4 crosstabulations of the data (e.g., white/nonwhite by the four age groups). With regard to sex, crash-involved motorcyclists were overwhelmingly male, although there were slightly more females among the older motorcyclists and especially those ages 35–44. Older motorcyclists were also more likely to be white, more likely to wear a helmet, and more likely to be seriously or fatally injured in their crash. Although the oldest riders (those ages 45+) were reported using alcohol or drugs at a slightly below average rate, alcohol/drug use was highest for those in the 35–44 year age group and lowest among 16–24 year-olds.

Table 1.

Motorcyclist Characteristic by Age of Operator (2000–2002 NC Data)

	Age of Motorcyclist
Motorcyclist Characteristic	16–24	25–34	35–44	45+	Overall
Male	97.51	95.5	91.1	94.2	94.7
White	76.3	64.5	80.6	91.0	77.2
Helmet worn	89.3	91.4	91.3	92.8	91.2
Serious (A+K) injury	14.6	16.1	19.3	21.2	17.6
Alcohol/drug use	4.8	8.1	11.8	7.2	7.9

¹Percent of all crash-involved motorcyclists in that age group.

Overall, two-thirds of the motorcycle crashes occurred on two-lane roadways, with only minor variations by age. However, there were significant (p<.0001) differences with respect to roadway type. Crashes involving older motorcyclists were more likely to occur on US or NC routes and on Interstate highways, whereas those involving younger motorcyclists were more likely to occur on the more minor secondary roadways and local streets (see Table 2).

Table 2.

Roadway Classification by Motorcyclist Age (2000–2002 NC data)

	Age of Motorcyclist
Road Class	16–24	25–34	35–44	45+	Overall
Interstate	2.31	3.5	4.0	4.4	3.5
US routes	15.0	12.5	16.0	22.5	16.2
NC routes	11.2	14.7	16.8	22.1	16.0
State secondary road	38.8	33.7	32.6	30.1	33.9
Local street	31.8	34.9	30.1	20.6	29.7
Other, including private property	0.9	0.8	0.5	0.4	0.7

¹Percent of all crash-involved motorcyclists in that age group.

Additional roadway/environmental variables are summarized in Table 3. Given the findings reported above, it is not surprising that crashes involving older motorcyclists were more likely to occur on higher speed roadways and in rural areas. They were also more likely to occur at intersection locations, on weekends, and during daytime/daylight hours. These results are all significant at p≤.01.

Table 3.

Roadway/environmental Characteristics of Motorcycle Crashes by Age of Operator (2000–2002 NC data).

	Age of Motorcyclist
Roadway/Environmental Characteristics	16–24	25–34	35–44	45+	Overall
Speed limit ≥50 mph	38.01	35.2	39.7	45.5	39.3
Rural	55.7	52.7	59.3	68.4	58.6
Weekend	39.8	41.9	44.5	44.3	42.5
Daytime (6am – 6pm)	56.7	58.7	63.2	75.7	63.2
Daylight	68.1	69.6	70.7	81.5	72.2
Intersection	27.6	29.6	31.4	33.0	30.3

¹Percent of all crash-involved motorcyclists in that age group.

Table 4 contains information on the first harmful event, i.e., the event that triggered the crash. This is a crash rather than a vehicle-oriented variable (i.e., only one such event is coded per crash), so that in crashes involving a motorcycle and one or more other vehicles, it may be associated with the motorcycle or with the other vehicle in the crash. Compared to younger motorcyclists, the crashes of older motorcyclists are less likely to be precipitated by running off the roadway; however, they are just as likely to involve an overturn or rollover in the roadway, or another non-collision event. For crashes where the first harmful event involved colliding with another vehicle, differences between older and younger riders were small.

Table 4.

First Harmful Event (Event Precipitating the Crash) by Age of Operator (2000–2002 NC Data)

	Age of Motorcyclist
First Harmful Event	16–24	25–34	35–44	45+	Overall
Ran off road	16.7	13.0	13.1	10.7	13.4
Overturn/rollover	19.2	19.6	15.7	20.5	18.8
Other non-collision	6.7	7.1	4.9	4.8	6.0
Fixed object	13.3	12.2	10.6	12.0	12.1
Rear-end slowing, stopping or turning	11.8	12.7	12.9	14.6	13.0
Left turn	8.6	11.4	13.5	10.9	11.1
Right turn	1.1	0.8	1.3	1.4	1.2
Head-on	1.9	1.7	1.9	1.3	1.7
Sideswipe	5.7	5.7	7.5	5.6	6.1
Angle	8.8	9.6	10.2	9.8	9.5
Other collision	6.2	6.2	8.4	8.4	7.1

¹Column percent.

In North Carolina, law enforcement officers are not asked to indicate which party is at fault in multi-vehicle crashes, and at-fault parties are not consistently cited for their role in the crash. However, officers are asked to identify specific driver/operator factors that they believe contributed to the crash. Overall, 60% of the motorcyclists had one or more contributing factors identified. The percentage was highest for riders ages 25–34 (71%), dropping to 62% for riders ages 25–34, and to 53–54% for riders ages 35–44 and 45+ (p<.0001).

Table 5 provides information on the distribution of “first” contributing factors, for those motorcyclists cited for one or more factors. Factors more likely to be cited for younger motorcyclists were exceeding the speed limit and careless, reckless or aggressive driving. Those more likely to be cited for older riders included inattention and alcohol involvement. These factors were all significant at p<.0001 when tested in 2×4 crosstabulations of the data (e.g., exceeding safe speed or not exceeding safe speed by age). Other factors were more equally distributed across the age categories. By far the most frequently cited factor was exceeding a safe speed.

Table 5.

Specific Factor Contributing to the Crash, When a Factor Was Identified, by Age of Motorcyclist (2000–2002 NC Data)

	Age of Motorcyclist
Driver Contributing Factors	16–24	25–34	35–44	45+	Overall
Exceeding safe speed	22.51	22.2	26.8	26.8	24.1
Exceeding speed limit	20.0	18.1	10.9	4.7	14.4
Failure to reduce speed	12.2	13.2	12.2	16.5	13.4
Careless, reckless or aggressive	12.5	9.8	4.9	2.2	8.1
Inattention	6.1	5.7	8.5	14.0	8.1
Swerved to avoid	3.7	6.3	4.1	5.8	5.0
Alcohol or drugs	1.6	3.2	6.9	4.8	3.8
Oversteered	3.2	3.1	3.9	3.4	3.3
Crossed center line	3.1	2.4	4.6	3.8	3.3
All other	15.1	16.0	17.2	18.0	16.5

¹Column percent.

SINGLE VERSUS MULTI-VEHICLE CRASHES

Just under half (49.1%) of the reported crashes were single vehicle events involving only the motorcyclist. The percentage of single vehicle crashes was highest among younger riders. Percentages were: 54.3% (ages 16–24); 50.5% (ages 25–34); 44.4% (ages 35–44); and 46.3% (ages 45+) (p<.0001).

Table 6 provides information on the characteristics of single and multi-vehicle motorcycle crashes. Within each age category, the table shows the percentage of all single vehicle, or all multi-vehicle, crashes involving a particular characteristic. With respect to alcohol/drug use, 11.9% of all single vehicle motorcycle crashes were found to involve a motorcyclist using alcohol or drugs, compared to a much lower 4.1% for multi-vehicle crashes. For motorcyclists aged 35–44, nearly one in five operators in motorcycle only crashes were judged to have been influenced by alcohol or drugs.

Table 6.

Characteristics of Single-vehicle and Multi-vehicle Crashes, by Age of Motorcyclist (2000–2002 NC Data).

		Age of Motorcyclist
Characteristic	Crash Type	16–24	25–34	35–44	45+	Over all
Alcohol/drug use	MC only	6.41	12.5	19.4	11.3	11.9
	2+ Veh	2.92	3.8	5.8	3.8	4.1
Estimated speed >60mph	MC only	22.3	18.3	11.1	6.6	15.4
	2+ Veh	9.1	7.7	4.2	1.9	5.7
Speed limit 50+ mph	MC only	47.8	44.5	51.0	57.0	49.5
	2+ Veh	26.4	25.9	30.7	35.6	29.6
≤ 2 lane roadway	MC only	81.2	81.6	87.1	81.3	82.5
	2+ Veh	53.8	50.9	53.6	59.0	54.2
Rural	MC only	67.5	66.2	73.9	83.0	71.8
	2+ Veh	41.6	39.0	47.7	55.8	45.8
Weekend	MC only	44.4	45.5	49.2	49.6	46.9
	2+ Veh	34.2	38.3	40.7	39.7	38.3
Daylight	MC only	65.0	64.5	65.5	77.2	67.7
	2+ Veh	71.7	74.8	74.9	85.2	76.7

¹Percent of all single-vehicle crashes involving that characteristic, for that age group.

²Percent of all two-vehicle crashes involving that characteristic, for that age group.

Single vehicle crashes also involved higher estimated traveling speeds, especially for riders in the two younger age categories. Motorcyclists ages 45+ were one-third as likely to be traveling 60 or more mph when involved in a single vehicle crash. Interestingly, although younger motorcyclists were traveling at generally higher speeds when they crashed, older motorcyclists were more likely than younger motorcyclists to be traveling on higher speed roadways. About a third of their multi-vehicle crashes, and over half of their single vehicle crashes, occurred on roadways with posted speed limits of 50 mph or greater.

Four out of five single vehicle crashes occurred on two-lane (or smaller) roadways, compared to just under half of multi-vehicle crashes, with even higher rates for riders ages 35–44. In addition, single vehicle crashes were more likely to occur in rural locations and during daylight hours. For both variables, the percentages increased with age. For riders age 45+, four of every five single vehicle crashes occurred on rural roadways and/or during daylight hours. Additional analysis showed that older motorcyclists were especially over-represented in single vehicle crashes occurring on busier US and NC roadways, and were under-represented in single vehicle crashes occurring on local streets.

DISCUSSION

Compared to automobiles, motorcycles are a high risk mode of travel. The overall motor vehicle fatality rate per 100 million vehicle miles traveled in the U.S. is 1.5; for motorcycles, it is 33.6 (NHTSA, 2003a; NHTSA 2003b). Moreover, the rate for all vehicle occupants has steadily declined, whereas the rate for motorcyclists is increasing. Given the trend towards increased motorcycle registrations, it is critical that steps be taken to improve motorcycle safety.

The results of the current study show that in North Carolina, as well as nationally, the recent increase in motorcyclist deaths primarily reflects an increase in crashes involving older riders. The number of riders ages 16–24 in crashes has, in fact, declined over the past decade, while the number of riders ages 25–34 has remained about the same. The sharpest increase in crash-involved motorcyclists has been those ages 45 and older. This trend has been noted elsewhere in the literature (NHTSA, February 2002); however, there has been little examination of the characteristics of older motorcyclist crashes directed at improving opportunities for intervention.

The analysis of the most recent three years of statewide NC crash data showed that older motorcyclists were more likely to be white, had a higher (but still low) proportion of females, and were more likely to wear a helmet. They were also more likely to be seriously injured in their crash, with one in five experiencing a fatal or incapacitating (A-level) injury. Part of this may be due to the location of their crashes, which tended to occur more often in rural areas and on higher speed roadways (US and NC routes). A separate analysis of crashes by region of the state showed that older motorcyclists were over-represented in crashes occurring in the western Mountain region of the state, while younger motorcyclists were over-represented in crashes occurring in the eastern Plains region. The rural western part of the state is a favorite destination of recreational riders, which would also help to explain the higher proportion of weekend and daytime riding by older motorcyclists.

Older motorcyclists were less likely than younger motorcyclists to be in run-off-road collisions, but they were about equally likely to overturn or roll over in the roadway. They were also less likely to be cited for contributing to their crash. Although older motorcyclists were somewhat more likely to be cited for exceeding a safe speed, they were five times less likely to be cited for exceeding the speed limit or for careless and reckless operation of their vehicle. And, even though their crashes were more likely to occur on higher speed roadways, their estimated travel speed was lower than for younger motorcyclists, in single vehicle as well as multi-vehicle crashes.

These results suggest that older riders are generally responsible riders, but with one exception: motorcyclists ages 35–44 had by far the highest rate of alcohol involvement in their crashes. This was especially true for single vehicle crashes, where one of five motorcyclists ages 35–44 was cited for alcohol or drug use.

Although involved in a somewhat lower percentage of single vehicle crashes overall, older motorcyclists were especially vulnerable to such crashes in rural areas and on busier US or NC roadways. While it is not possible, without adequate exposure data, to draw conclusions about the riskiness of such riding, it is safe to say that the effectiveness of education and enforcement programs might be increased if they targeted these types of higher speed, rural settings. And, of course, they should target alcohol use by older motorcyclists.

It remains unclear as to why older motorcyclists are more likely to be seriously injured or killed in crashes. Certainly increasing age brings with it a greater degree of fragility, so that in a given crash, an older motorcyclist is more likely to die than a younger motorcyclist. But given that most of the “older” motorcyclists are under the age of 60, it seems unlikely that this is the sole explanation for the nearly 50% higher risk of serious injury observed in the NC data. Again, it may be tied to their exposure on higher speed roadways and in rural areas, where access to emergency medical care may be more limited.

A weakness of the current investigation is the absence of exposure data more detailed than number of registered vehicles. In particular, information is needed on where motorcyclists ride, for what purposes, on what types of motorcycles, and most importantly, for what distances. A recent Australian study funded by the Motor Accidents Authority of New South Wales surveyed a sample of registered motorcycle owners and then re-surveyed the same sample six months later to capture paired odometer readings for use in estimating annual kilometers ridden (Christie, 2003). However, other than information collected every few years by the Motorcycle Industry Council as part of its Motorcycle Owner Survey (MIC, 1998), little is known about the riding habits of U.S. motorcyclists.

As part of the current project, a random sample of registered motorcycle owners in North Carolina is being surveyed. The survey will gather information on nature and extent of riding and years of riding experience, and will be linked to information about the motorcycle from the vehicle registration file (engine size, model year, etc.). In addition, both the survey and the crash data will be linked to driver history data to provide information regarding license status of the motorcycle operator and length of licensure. By combining data from a number of sources, we hope to develop a clearer understanding of factors impacting motorcyclist safety.

As an example, one important question that needs addressing is whether the observed decrease in crashes per registered motorcycle is due to the greater proportion of older, and presumably safer, riders, or simply to a decrease in the annual distances these riders are traveling. It will also be of interest to learn more about these “older” riders, for example, how many are long term riders versus returning or new riders?

Given the expected continued growth in motorcycle sales it will be important to continue to look for ways to promote motorcyclist safety. Combining new approaches targeting older riders with more traditional approaches such as rider training and motorist education, will be needed. In addition, it is critical that the trend of repealing or weakening universal helmet laws be reversed.