Size and age of fatal drivers by crash type, vehicle type and gender

Abstract Objective: The objective of this study was to determine the physical characteristics of fatal drivers in motor vehicle crashes with focus on rear impacts. Methods: 1998 to 2020 FARS data was analyzed for height, weight, and age of fatal drivers. The data was queried by gender, crash type and vehicle type. Results: The average fatal driver weighed 80.4 kg, was 173.4 cm tall, and was 43 years old. Females were 16.0 kg lighter and 14.2 cm shorter than males on average. The height was 151.2 cm for the 5th percentile female, 177.0 cm for the 50th male and 188.9 cm for the 95th male. The weight of fatal drivers increased linearly with calendar year. The increase rate was greater in females than in males. About 3% of fatal drivers were involved in rear crashes, 39.9% in frontal crashes and 36.8% in rollovers. The average fatal driver was 172.5 cm tall and weighed 81.0 kg in rear impacts. They were similar in height and weight to the overall sample. The average fatal driver in rear impacts was 46 years old, 3 years older than the overall average. Pickup truck drivers weighed 85.4 kg and were 176.8 cm tall on average. They were heavier and taller than passenger car drivers on average, which were 78.0 kg and 172.2 cm. Fatally injured minivan drivers were 10 years older than fatally injured passenger car drivers on average. The findings are compared with ATDs (anthropometric test devices) used in sled and crash testing. Conclusion: The average weight of fatal drivers increased with calendar year. The average size of fatal drivers was similar by crash types. Fatal drivers were older in rear impacts.


Introduction
Seats are key component of occupant safety in rear impacts. Seat performance is evaluated in crash and sled testing using Anthropomorphic Test Devices (ATDs), commonly referred to as "crash test dummies" or "dummies". Most rear tests are conducted with ATDs representing a 50 th percentile male. For example, NHTSA uses the Hybrid III 50 th male in FMVSS 202a and IIHS uses the BioRID II to evaluate head restraints in rear sled tests (NHTSA 2010, IIHS 2020. Other ATDs have been used to evaluate seat performance, such as the THOR-50M, in moderate to severe rear tests (Hagedorn et al. 2022).
The Hybrid III 50 th male is the most widely used ATD. It was developed by General Motors under contract from the National Highway Traffic Safety Administration (NHTSA) (Foster et al. 1977). The ATD size was designed to be representative of the US population in the 1970s (Reed and Rupp 2013). The segment size and center of gravity locations were based on anthropometric guidelines (Foster et al. 1977) and SAE J963 specifications (Schneider et al. 1983).
Most drivers in the 1960s were males (Sivak 2013, Sivak 2015. Males also drove more miles than females at the time and consequently were more often involved in vehicle crashes (Popkin et al. 1986). For example, the Federal Highway Administration reported in 1985 that males drove approximately twice as many miles as females. Today, there are an approximately equal number of male and female drivers, so testing with female ATD has been added (Sivak 2015).
Various ATD sizes have been developed based on gender and size (Mertz et al. 1989). The Hybrid III 5 th female was created to represent smaller size occupants. Mertz et al. (2001) reported that the size of the Hybrid III 5 th female was similar to a 13-year-old teenager. Other sizes of ATDs have been developed, including a Hybrid III 95 th male and various child ATDs.
Fatality Analysis Reporting System (FARS) is a publicly available database. It is a census of fatal motor vehicle crashes in the 50 States, District of Columbia, and Puerto Rico. Since 1998, FARS has reported driver height and weight. One aim of this study was to query the database to identify the size of fatal drivers in all and rear crashes. The data was analyzed by age and gender and different vehicle types. The average and 5 th , 50 th , and 95 th percentile data for fatal drivers was assessed with respect to weight and height. The results can be used to assess whether ATD sizes representative of fatally injured US drivers and to the US population. The FARS data was also analyzed by calendar year since the average body weight of the US population has increased over time Rupp 2013, Fryar et al. 2018).
A summary table was provided in the supplemental material to facilitate comparisons between the FARS and US population (Fryar et al. 2016(Fryar et al. , 2021 and ATD data (see Appendix A, Table A1). For example, Fryar et al. (2016) reported the average height and weight for men and women from 2011 to 2014 ("population 11-14" in Table A1). In a later study , average height and weight information was reported for men and women from 1999 to 2016. The data was reported in two calendar year increments. The results indicate that the average weight increased by 3.9 kg (85.9 vs. 89.8 kg) for adult males and by 3.  Fryar et al. (2021) reported the height and weight data by percentile for male and female individuals. The data is referred to as "population 15-18" in Table A1. Table A1 also shows the height and weight by gender for ATDs nominally representing 5 th , 50 th , and 95 th percentile adults. It focused on the most commonly used adult ATDs currently available. The height information for the Hybrid 5 th female, 50 th male and 95 th male was based on values reported in Mertz et al. (2001). The height for the BioRID was obtained from Sato et al. (2017). The weight information was obtained from Humanetics (https://humanetics.humaneticsgroup.com/ products/anthropomorphic-test-devices).

FARS
FARS is the Fatality Analysis Reporting System (www.nhtsa. dot.gov). FARS data for calendar years 1998-2020 was evaluated for light vehicles. Light vehicles included passenger cars (BODYTYPE ¼1-8,17), SUVs (BODYTYPE ¼14-16, 19), minivans (BODYTYPE ¼20), and pickup (PU) trucks (BODYTYPE ¼30-39) with model year 1994 or greater. Drivers were identified as occupants sitting in the left-front seat (SEAT_POS ¼11) and with fatality (INJ_SEV ¼ 4). Drivers were divided into males (SEX ¼1) and females (SEX ¼2). Weight was defined using the driver weight variable (17.7 < DR_WGT < 318.0 kg), height using driver height (58.4 < DR_HGT < 274.3 cm) and age (14 < age < 99 yo). The limits were selected to remove unknowns, possible coding errors and small children sitting on the driver's seat. There were 214,316 fatal drivers based on the age, height and weight inclusion criteria including vehicle type, model year, seating location, age, height and weight.
Crash types included frontal impacts (IMPACT1 ¼ 1,11,12 and ROLLOVER< ¼0), near-side impacts (IMPACT1 ¼ 8,9,10 and ROLLOVER < ¼0), far-side impacts (IMPACT1 ¼ 2,3,4 and ROLLOVER < ¼0), rear impacts (IMPACT1 ¼ 5,6,7 and ROLLOVER < ¼0), and rollovers (ROLLOVER >0). The mean, standard errors and percentiles were determined using the PROC SURVEYMEANS in SAS. A t-test was used to compare means. Table 1 summarizes the average height, weight and gender for fatal drivers in the 1998 to 2020 FARS for all and rear crashes based on the inclusion criteria. Overall, FARS included 225,255 fatal drivers; 72,015 were females and 153,205 were males. On average, fatal drivers weighed 80.4 kg and were 173.4 cm tall. Females were 16.0 kg lighter and 14.2 cm shorter than males. There were 6,724 fatal drivers in rear crashes; 39% of which were females. The average height and weight in rear crashes was similar to all crashes. Table 1 also summarizes the average age in all and in rear crashes. The average fatal driver was 46.3 years old in rear crashes and 43.4 years old in all crashes.

Results
Appendices A-F provide additional information including the age, height, and weight by average, minimum, maximum, and standard deviation. Table A1 provides a summary of the FARS data. Compared to the US population, fatally injured male drivers in the '98-'20 FARS data were 0.4 kg lighter and 2.3 cm taller than the average male in 1999-2000 and 5.0 kg lighter and 2.4 cm taller than the average male in 2015-2018. Fatally injured females were 4.8 kg and 8.0 kg lighter and 1.7 cm and 2.0 cm taller for the same corresponding US population. Table A1 also summarizes the 5 th , 50 th , and 95 th percentile fatal driver and the US population in 2011 to 2014 (Fryar et al. 2016) and 2015 to 2018 (Fryar et al. 2021). Overall, the 5 th percentile weight of fatal drivers was 52.5 kg, the 50 th percentile was 77.1 kg, and the 95 th percentile was 117.5 kg. The corresponding heights were 155.2 cm, 172.7 cm, and 187.5 cm. For males, the 95 th percentile fatal driver weight was 120.1 kg, and the height was 188.9 cm in the 1998-2020 FARS data. Compared to the 2015-2018 US population, the 5 th percentile female in the 1998-2020 FARS was 2.1 kg lighter and 1.4 cm taller. For the 50 th percentile data, the fatally injured female and male were 0.9 cm and 1.6 cm taller than the average 2015-2018 U.S. population, respectively. The female was 7.6 kg lighter, and the male was 5.8 kg lighter.   Figure 3 shows the average weight of fatal drivers by calendar year and gender. The weight increased linearly with calendar year. The increase rate was greater in fatal females than in males. Appendix D provides additional weight information including average, minimum, maximum, and standard deviation. On average, fatal drivers were 43.4 years (yr) old. The average age for fatal females was 42.9 years old (5 th percentile: 17.3 yr and 95 th percentile: 80.5 yr). It was 44.8 years old (5 th percentile: 17.8 yr and 95 th percentile: 80.1 yr) for males. Figure 4 shows the average age, weight, and height by crash type. Figure F1 shows the corresponding distribution of fatal drivers. About 3.0% of fatal drivers were involved in rear crashes, 40.2% in frontal crashes and 34.6% in rollovers ( Figure F1). The average fatal driver was 172.5 cm tall and weighed 81.0 kg in rear impacts. They were similar in height and weight to the overall sample. The average fatal driver in rear impacts was 46 years old, 3 years older than the overall average. Figure 5 shows the average age, height, and weight of fatal drivers by vehicle type. Minivan fatal drivers were older than passenger car drivers, at 52.0 and 42.6 respectively. Pickup (PU) truck fatal drivers were heavier and taller, at 85.4 kg and 176.8 cm, than passenger car drivers, at 78.0 kg and 172.2 cm (p < 0.0001). Appendix E provides additional information including standard deviations and errors.
Appendix G tabulates the age, weight, and height data by average, minimum, maximum, standard deviation and error, and sample size for fatal drivers in pickup trucks. The data is shown by gender. There were 43,795 males and 5,772 females. The higher proportion of male drivers than female drivers (88.3% v 11.6%) may explain the higher driver height in pickup trucks than in other types of vehicles. Figure G1 shows the average age, weight, and height of fatal     drivers in all crashes with pickup trucks. On average, male drivers were 13.9 cm taller than female drivers and 18.0 kg heavier.

Discussion
The size and age of fatal drivers was assessed by crash type to identify if there were different needs by impact direction. The distribution of fatal drivers by crash type was frontal crashes (40.2%), rollover (34.6%) with about 3% in rear crashes. The height and weight of fatal drivers were similar, irrespective of crash type. On average, fatal drivers weighed 80.4 kg and were 173.4 cm tall. These results suggest that the size of current 50 th percentile male ATDs is representative of the average overall fatal driver in motor vehicle crashes. The 50 th percentile ATDs include 50 th Hybrid III, BioRID and THOR-50M used to evaluate seat performance in rear crashes.
The height and weight for the Hybrid III 5 th female is similar to the average 5 th percentile female fatal driver in FARS. The SID-IIs was 3.8 kg lighter. The Hybrid III 50 th male, THOR-50M, and the BioRID ATDs have a weight of about 76-78 kg, and the Hybrid III 50 th male has a height of 175.1 cm as reported by Mertz et al. (2001). The results indicate these ATDs are representative of the average 50 th percentile fatal drivers, irrespective of gender. The THOR-AV-50M is 8.5 kg heavier than the THOR-50M; its weight is more closely comparable to the 50 th percentile fatally injured male driver. The Hybrid III 95 th male is 186.4 cm, 2.5 cm shorter than the 95 th percentile male fatal driver in FARS. The ATD is 18.9 kg lighter than the 95 th percentile fatally injured male driver. It is 2.9 kg lighter than the 95 th percentile fatally injured female driver. Reed and Rupp (2013) compared the size of ATDs to 2005-2008 NHANES (National Health and Nutrition Examination Survey) data. They reported that the weight of the Hybrid III 50 th male represented the 33 rd percentile body weight for U.S. adults. The authors showed that the weight of a 50 th percentile male in the U.S. in 2008 was 85.4 kg, similar to the THOR-AV-50M at 85.1 kg (less than 5% difference).
The demographics of fatal drivers in 1998-2020 FARS was also compared to the U.S. population. Fatally injured females were 4.8 kg and 8.0 kg lighter than the U.S. population in 1999-2000 and in 2015-2018, respectively. Their height was within 5% of the 5 th female in the same U.S. populations. Fatally injured male drivers were 0.4 kg to 5.0 kg lighter and about 2.4 cm shorter. According to CDC (Fryar et al. 2021), a height of 187.4 cm still corresponds to the 95 th percentile height of males today.
The FARS data showed that weight of fatally injured drivers has increased with time. The U.S. population's average weight has also increased over time. Reed and Rupp (2013) showed a relatively small increase in weight from 1974 to 2008 for the 5 th percentile U.S. female compared to the 50 th percentile U.S. female. A weight of 100 kg is approximately representative of a 75 th percentile male.
The effect of occupant size and age was assessed by vehicle type. Fatal pickup truck drivers were taller and heavier than drivers of other vehicles (p < 0.0001). Most fatal pickup truck drivers were males. On average, men are taller than women. Pickup trucks are high-riding vehicles, and it may be more difficult for shorter individuals to get in and out of the vehicle. Fatal minivan drivers were older than fatal passenger car drivers, at 52.0 and 42.6 years, respectively (p < 0.0001). Minivans generally have higher occupancy than other types of vehicles. It may be easier for children and older rear seat occupants to get in and out with the sliding doors (Lu et al. 2016), making this type attractive to older drivers.
Current ATDs have varying representativeness to the populations in different countries. Cao et al. (2016) compared the anthropometric data of ATDs and Chinese adults. They reported that in 2010, the median (50 th percentile) Chinese males were about 3.48% shorter and weighed 11.89% less than the Hybrid III 50 th male ATD. Similarly, Chinese females were about 2.06% lower in stature and about 8.03% lower in body weight than the ATD.
Limitations: First, the height and weight of fatally injured occupant information in FARS was likely obtained from the driver's license, where height and weight are self-reported. Studies have shown inaccuracies in self-reported height and weight data (Fillenbaum et al. 2010).
Second, the average weight of fatally injured females was 8.2 kg lighter than the average weight in the 2015-2018 population (Fryar et al. 2021). The average height and weight of fatal drivers was assessed using 1998 to 2020 FARS data. An additional analysis was conducted by grouping the data into [1998][1999][2000][2001][2002][2003][2004][2005][2006][2007][2008][2009][2010][2011] and 2012-2020 calendar years to assess whether height and weight varied over time. The results are summarized in Table A1 and showed that the average weight increased with time irrespective of gender while height did not.
Third, there are various weight discrepancies observed in reported ATD specifications. For example, the Hybrid III 5 th female weighs 49 kg according to NHTSA and Humanetics. The weight is listed as 47 kg in Mertz et al. (2001). The standing height of the ATDs are not often listed. The ATDs used in regulatory testing by NHTSA are designed to represent seated occupants. The Hybrid III 50th male was designed to represent a 175 cm occupant (Reed and Rupp 2013). The equivalent standing height may be shorter since it is difficult to lay the ATD flat on its back.
Fourth, the physical dimensions of fatally injured male, female and all drivers were determined, irrespective of crash type and restraint use. The risk for injury and outcome was not assessed. Research has been conducted to assess the effect of occupant sizes on kinematics and biomechanical responses (Carter et al. 2014, Zhang et al. 2017. The results are leading to the development of new occupant sizes, shapes and weights for computational studies (Zhang et al. 2017).

Funding
The author(s) reported there is no funding associated with the work featured in this article.