Comparison of work-related traffic crashes between male taxi drivers aged ≥65 years and <65 years in South Korea

Abstract

BACKGROUND:

The percentage of drivers aged ≥65 years among all Korean taxi drivers has risen sharply from 3.2% in 2006 to 22.0% in 2016.

OBJECTIVE:

This study compared the characteristics of work-related traffic crashes between male taxi drivers aged ≥65 years and <65 years.

METHODS:

Using the national compensation data of South Korea, 586 injured male taxi drivers were analyzed based on driver-related (work experience, company size, employment status, injured body part, and injury type) and crash-related factors (time and day of the crash, weather condition, road type, violation and drowsiness) by age group.

RESULTS:

For the injured drivers aged ≥65, percentages of the crashes related to some factors were lower than those of the drivers aged <65 years: at night (55.2% vs. 64.0%), on rainy or snowy day (7.8% vs. 21.3%), on straight road (40.5% vs 50.9%) and on the expressway (6.9% vs. 13.0%). However, the percentage of the crashes related to a violation for the injured drivers aged ≥65 years (23.3%) was higher than that of the drivers aged <65 years (13.4%). Furthermore, the taxi drivers aged ≥65 years had a higher death rate (14.7%) than the drivers aged <65 years (8.5%).

CONCLUSIONS:

The results can be useful for injury prevention policies and guidelines for elderly taxi drivers such as strengthened qualification tests for the aged drivers and improvement of the working environment.

Keywords

Occupational injury elderly male driver-related factors. crash-related factors taxi drivers

1 Introduction

Korean Standard Classification of Occupations defines that taxi drivers are those who drive taxis to transport passengers quickly and safely to their destination [1]. In 2016, there were 166,096 companies and 284,920 employees, including 164,468 self-employed workers in the taxi transportation industry of South Korea [2]. In the USA, taxi drivers, ride-hailing drivers, and chauffeurs held about 305,100 jobs. 36% of them were self-employed workers, and 16% of them were employed by taxi and limousine service in 2016 [3].

According to the Occupational Outlook Handbook, taxi drivers, ride-hailing drivers, and chauffeurs have one of the highest rates of injuries and illnesses of all occupations, and most injuries result from traffic crashes [3]. About 1 in 4 worked part-time and evening and weekend work is common while some drivers work late at night or early in the morning [3].

Professional drivers are high-risk workers and comprise nearly one-third of all work-related deaths [4]. Comparison of mortality for specific and general categories of occupations shows that high-risk specific occupations include taxi drivers, cooks, longshoremen, and transportation operatives [5].

In Korea, attention to the safety of taxi drivers is essential because there are many taxi drivers and a taxi is one of the major transportation vehicles for the commuters.

Moreover, The percentage of the older taxi driver (≥65 years old) among all Korean taxi drivers has risen sharply from 3.2% in 2006 to 22.0% in 2016, which is three times larger than those of bus (6.6%) and truck (7.5%) drivers [6]. As in many developed countries [7], many people in Korea tend to find occupations beyond the current retirement age, and there are lower entry barriers for taxi drivers than for bus or truck drivers. Among all the crashes incurred by taxi, the proportion of crashes caused by older taxi drivers increased to 18.4% while the proportion of deaths caused by older taxi drivers among all the deaths incurred by taxi also increased to 21.3% in 2015 [6]. According to the previous studies, older taxi drivers have fewer crashes than other age groups, but there is an increased risk of death and severe injury in the event of a crash [8 –15].

Among 119,997 corporate-employed taxi drivers in 2016, 97.0% were male [2]. Also, 97.4% of taxi drivers causing the crashes during 2013–17 were male [16].

The crash rate of a corporation taxi is higher than that of a self-employed taxi because of frequent daytime and nighttime shift change [17]. Therefore, we focus on the crash characteristics of older male drivers who work for corporations.

The working environment of a taxi driver, work shift patterns, and breaks affect crash characteristics [17]. Traffic crashes are a major cause of health problems or premature death in developing countries. The driver’s age, license type, employment status, income level, the use of seat-belt has a significant association with traffic crashes [18]. Age-related declines in cognitive, perceptual, and motor capabilities affect driving performance negatively [19]. Flower et al. [20] summarized the researches on cognitive function between younger and older drivers, and Varianou-Mikellidou et al. [21] addressed occupational health and safety management for aging workforces and discussed factors affecting working ability and worker’s performance. Newnam et al. [22] explored the differences in driving behavior across age and years of education of taxi drivers. Lam [23] investigated the associations between some environmental factors (day of the crash, nighttime, weather condition, road shape, carrying passengers) and the increased risk of motor vehicle crash-related injuries among taxi drivers.

Also, qualitative researches for crash characteristics as well as the behavior of taxi drivers are needed to prevent the traffic crashes caused by taxi drivers. Some researchers investigate the aspects of crashes on the driving habits, violation, drowsiness, and fatigue of taxi drivers [24, 25].

This study compared the characteristics of work-related traffic crashes of male taxi drivers between those aged ≥65 years and <65 years. By analyzing the national compensation data of South Korea, this study investigates the injured male drivers employed by taxi companies based on driver-related factors (work experience, company size, employment status, injured body part, and injury type) and crash-related factors (time and day of the crash, weather condition, road type, violation, and drowsiness) by age group. The injured male driver aged 65 or over, is the main concern of the analysis.

2. Methods

2.1 Data collection

From the national compensation data of South Korea, this study analyzes 586 injured male taxi drivers approved as industrial compensation accidents requiring four days or more sick leave. Of the 8,173 occupational injuries in the transportation and warehousing industry during 2015 –2016, 1,840 injuries were drivers due to traffic crashes. Among them, 616 injuries were taxi drivers. Of 616 injured taxi drivers, 15 self-employed and 15 female drivers were excluded.

2.2 Research variables and data analysis

This study investigated driver-related factors and crash-related factors, as shown in Table 1, using the industrial accident report. Driver-related factors include work experience, company size (size of employment), employment status, injured body part, and injury type. Crash-related factors involve the day of the week, time of the day and weather conditions at the time of clash, road type, and road shape of crash location, violation, and drowsiness of the driver at the time of the crash.

Table 1
Definition of factors and variables of this study

Factors Variable Description

Injured driver

Injury severity 1, nonfatal injury; 2, death

Work experience(years) 1, under 1yr; 2, 1–3yrs; 3, 3–10yrs; 4, over 10yrs

Company size (number of employees) 1, under 30; 2, 30–99; 3, 100–199; 4, 200 or more

Employment status 1, regular; 2, temporary

Injured body part 1, facial; 2, neck; 3, shoulder; 4, chest/back; 5, upper limb; 6, lower limb; 7, complex

Injury type 1, cerebral hemorrhage;

2, fracture/crushing/dislocation;

3, sprain; 4, rupture; 5, contusion/laceration/abrasion;

6, complex

Crash

Day of the week 1, weekdays(Mon-Fri); 0, weekend(Sat-Sun)

Time of day 1, daytime; 0, nighttime (p.m. 8:00–a.m. 5:00)

Weather condition 1, rainy or snowy; 0, otherwise

Road type 1, expressway; 0, general road (rural road or urban street)

Road shape of accident location Intersection: 1, if intersection; 0, otherwise

Straight: 1, if straight; 0, otherwise

Curved: 1, if curved; 0, otherwise

Downhill: 1, if downhill; 0, otherwise

Tunnel: 1, if tunnel, 0, otherwise

Violation 1, with violation; 0, without violation

Drowsiness 1, yes; 0, no

Factors	Variable	Description
Injured driver
	Injury severity	1, nonfatal injury; 2, death
	Work experience(years)	1, under 1yr; 2, 1–3yrs; 3, 3–10yrs; 4, over 10yrs
	Company size (number of employees)	1, under 30; 2, 30–99; 3, 100–199; 4, 200 or more
	Employment status	1, regular; 2, temporary
	Injured body part	1, facial; 2, neck; 3, shoulder; 4, chest/back; 5, upper limb; 6, lower limb; 7, complex
	Injury type	1, cerebral hemorrhage;
		2, fracture/crushing/dislocation;
		3, sprain; 4, rupture; 5, contusion/laceration/abrasion;
		6, complex
Crash
	Day of the week	1, weekdays(Mon-Fri); 0, weekend(Sat-Sun)
	Time of day	1, daytime; 0, nighttime (p.m. 8:00–a.m. 5:00)
	Weather condition	1, rainy or snowy; 0, otherwise
	Road type	1, expressway; 0, general road (rural road or urban street)
	Road shape of accident location	Intersection: 1, if intersection; 0, otherwise
		Straight: 1, if straight; 0, otherwise
		Curved: 1, if curved; 0, otherwise
		Downhill: 1, if downhill; 0, otherwise
		Tunnel: 1, if tunnel, 0, otherwise
	Violation	1, with violation; 0, without violation
	Drowsiness	1, yes; 0, no

Among the variables, regular worker has a term of employment of one year or more, and a temporary worker is a temporary or daily worker who has a duration of employment of less than one year. Road types are classified into expressways that are exclusively for cars and general roads (rural road or urban street) where traffic lights are present. Injured body part refers to the significant injured body part, and complex refers to injuries to the whole body. A violation means a crash caused by intentional offense such as signal violation, centerline violation, illegal U-turn, and lane violation except speeding.

In this study, we divided the total injured male drivers into the older drivers who were ≥65 years old and the drivers aged <65 years. For each age group and crash factor, the analysis is made for the total injured drivers (including fatal and non-fatal) and the dead drivers respectively. For the frequency distribution of age group, the χ² test was done using SPSS with p values below 0.05 as statistically significant.

3 Results

3.1 Characteristics of the injured drivers in view of driver-related factors

3.1.1 Analysis of injured drivers by severity

Table 2 shows the distribution of 586 injured drivers according to the severity of the injury. Among 586 casualties, 529 (90.3%) were non-fatal, and 57 (9.7%) were fatal. 470 (80.2%) of the total injured were under 65 years old, and 116 (19.8%) were aged 65 years or older. There was a statistically significant difference in the distribution of injuries between the age groups (χ² = 4.000, p = 0.038). The percentage of the dead among the injured who were aged <65 years was 8.5% while the mortality rate of those aged ≥65 years was 14.7%.

Table 2
Distribution of injured drivers by severity

Injury severity Age(years) Total

<65 ≥65

Non-fatal injury 430 99 529

91.5% 85.3% 90.3%

Death 40 17 57

8.5% 14.7% 9.7%

Total 470 116 586

100.0% 100.0% 100.0%

80.2% 19.8% 100.0%

χ² test χ² = 4.000, p = 0.038

Injury severity	Age(years)	Total
Non-fatal injury	430	99	529
	91.5%	85.3%	90.3%
Death	40	17	57
	8.5%	14.7%	9.7%
Total	470	116	586
	100.0%	100.0%	100.0%
	80.2%	19.8%	100.0%
χ² test	χ² = 4.000, p = 0.038

3.1.2 Analysis of injured drivers by work experience

Table 3 shows the distribution of 586 injured drivers (including fatal and non-fatal) by work experience. Among 586 casualties, 240 (41.0%) had less than one year of work experience and, overall, 378 (64.5%) had less than three years of work experience. There was no difference between the age groups by the work experience (χ² = 4.413, p = 0.220).

Table 3
Distributions of injured drivers by work experience

Work experience (years) Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

<1 200 40 240 9 6 15

42.6% 34.5% 41.0% 4.5% 15.0% 6.3%

1–3 112 26 138 13 5 18

23.8% 22.4% 23.5% 11.6% 19.2% 13.0%

3–10 121 36 157 13 5 18

25.7% 31.0% 26.8% 10.7% 13.9% 11.5%

≥10 37 14 51 5 1 6

7.9% 12.1% 8.7% 13.5% 7.1% 11.8%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 4.413, p = 0.220 χ² = 1.310, p = 0.72

Work experience (years)	Injury (fatal and nonfatal)	Death
<1	200	40	240	9	6	15
	42.6%	34.5%	41.0%	4.5%	15.0%	6.3%
1–3	112	26	138	13	5	18
	23.8%	22.4%	23.5%	11.6%	19.2%	13.0%
3–10	121	36	157	13	5	18
	25.7%	31.0%	26.8%	10.7%	13.9%	11.5%
≥10	37	14	51	5	1	6
	7.9%	12.1%	8.7%	13.5%	7.1%	11.8%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 4.413, p = 0.220	χ² = 1.310, p = 0.72

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by the work experience (χ² = 1.310, p = 0.727). The mortality rate of the injured drivers with less than one year of work experience was 6.3% which is calculated by dividing 15 (total number of dead drivers who had less than one year of work experience) with 240 (total number of injured drivers who had less than one year of work experience). The mortality rates of other work experience groups were 11.5–13.0%. The mortality rates of those aged ≥65 years are higher than the mortality rates of those aged <65 years except for those having work experience of more than ten years.

3.1.3 Analysis of injured drivers by the company size

Table 4 presents the distribution of 586 injured drivers (including fatal and non-fatal) by the company size. Among 586 casualties, 252 (43.0%) worked for companies with 100–199 employees and 208 (35.5%) work for companies with 30–99 employees. There was no difference between the age groups by the size of employment (χ² = 3.222, p = 0.359).

Table 4
Distributions of injured drivers by company size

Size of employment Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

<30 45 9 54 3 2 5

9.6% 7.8% 9.2% 6.7% 22.2% 9.3%

30–99 168 40 208 16 3 19

35.7% 34.5% 35.5% 9.5% 7.5% 9.1%

100–199 195 57 252 14 9 23

41.5% 49.1% 43.0% 7.2% 15.8% 9.1%

≥200 62 10 72 7 3 10

13.2% 8.6% 12.3% 11.3% 30.0% 13.9%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 3.222, p = 0.359 χ² = 2.987, p = 0.394

Size of employment	Injury (fatal and nonfatal)	Death
<30	45	9	54	3	2	5
	9.6%	7.8%	9.2%	6.7%	22.2%	9.3%
30–99	168	40	208	16	3	19
	35.7%	34.5%	35.5%	9.5%	7.5%	9.1%
100–199	195	57	252	14	9	23
	41.5%	49.1%	43.0%	7.2%	15.8%	9.1%
≥200	62	10	72	7	3	10
	13.2%	8.6%	12.3%	11.3%	30.0%	13.9%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 3.222, p = 0.359	χ² = 2.987, p = 0.394

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by the company size (χ² = 2.987, p = 0.394). The mortality rate of the injured drivers employed by companies with more than 200 employees was 13.9% (10/72) which was higher than those of other size groups. Except for those who worked for the companies with 30–99 employees, the mortality rates of those aged ≥65 years were much higher than the mortality rates of those aged <65 years.

3.1.4 Analysis of injured drivers by employment status

Table 5 gives the distribution of 586 injured drivers by employment status. Among 586 casualties, 513 (87,5%) were regular workers, and 73 (12.5%) were temporary workers. There was no difference between the age groups by the employment status (χ² = 2.040, p = 0.104).

Table 5
Distribution of injured drivers by employment status

Employment status Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Regular 416 97 513 34 14 48

88.5% 83.6% 87.5% 8.2% 14.4% 9.4%

Temporary 54 19 73 6 3 9

11.5% 16.4% 12.5% 11.1% 15.8% 12.3%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 2.040, p = 0.104 χ² = 0.063, p = 0.542

Employment status	Injury (fatal and nonfatal)	Death
Regular	416	97	513	34	14	48
	88.5%	83.6%	87.5%	8.2%	14.4%	9.4%
Temporary	54	19	73	6	3	9
	11.5%	16.4%	12.5%	11.1%	15.8%	12.3%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 2.040, p = 0.104	χ² = 0.063, p = 0.542

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by the employment status (χ² = 0.063, p =0.542). The mortality rate of the injured drivers who were regular workers was 9.4% (48/513) while the mortality rate of the injured drivers who were temporary workers was 12.3%. Irrespective of the employment status, the mortality rates of those aged ≥65 years were much higher than the mortality rates of those aged <65 years.

3.1.5 Analysis of injured drivers by injured body part

Table 6 shows the distribution of 586 injured drivers by the injured body parts. Among 586 casualties, the proportions of the injured body parts were in the following order: 226 (38.6%) on the chest/back, 125 (21.3%) on the face/head, 123 (21.0%) on the neck. There was no difference between the age groups by the injured body part (χ² = 5.982, p = 0.425).

Table 6
Distribution of injured drivers by injured body part

Injury body part Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Face/head 99 26 125 16 5 21

21.1% 22.4% 21.3% 16.2% 19.2% 16.8%

Neck 99 24 123 2 2 4

21.1% 20.7% 21.0% 2.0% 8.3% 3.3%

Shoulder 31 5 36 – – –

6.6% 4.3% 6.1% – – –

Chest/back 174 52 226 18 9 27

37.0% 44.8% 38.6% 10.3% 17.3% 11.9%

Upper limb 21 4 25 – – –

4.5% 3.4% 4.3% – – –

Lower limb 41 4 45 2 0 2

8.7% 3.4% 7.7% 4.9% 0.0% 4.4%

Complex 5 1 6 2 1 3

1.1% .9% 1.0% 40.0% 100.0% 50.0%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 5.982, p = 0.425 χ² = 2.167, p = 0.705

Injury body part	Injury (fatal and nonfatal)	Death
Face/head	99	26	125	16	5	21
	21.1%	22.4%	21.3%	16.2%	19.2%	16.8%
Neck	99	24	123	2	2	4
	21.1%	20.7%	21.0%	2.0%	8.3%	3.3%
Shoulder	31	5	36	–	–	–
	6.6%	4.3%	6.1%	–	–	–
Chest/back	174	52	226	18	9	27
	37.0%	44.8%	38.6%	10.3%	17.3%	11.9%
Upper limb	21	4	25	–	–	–
	4.5%	3.4%	4.3%	–	–	–
Lower limb	41	4	45	2	0	2
	8.7%	3.4%	7.7%	4.9%	0.0%	4.4%
Complex	5	1	6	2	1	3
	1.1%	.9%	1.0%	40.0%	100.0%	50.0%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 5.982, p = 0.425	χ² = 2.167, p = 0.705

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by the injured body part (χ² = 2.167, p =0.705). The mortality rate of the injured drivers who injured on the face/head was 16.8%, followed by the chest/back (11.9%). Notably, the mortality rate of the injured drivers who injured all over the body (complex) was as high as 50.0%. Except for those who injured on the lower limb, the mortality rates of those aged ≥65 years were much higher than the mortality rates of those aged <65 years.

3.1.6 Analysis of injured drivers by injury type

Table 7 gives the distribution of 586 injured drivers by injury type. Among 586 casualties, the proportions of the injury type were in the following order: 294 (50.2%) by fracture/crushing/ dislocation, 129 (22.0%) by a sprain, 70 (11.9%) by a cerebral hemorrhage. There was no difference between the age groups by injury type χ² = 4.093, p = 0.536).

Table 7
Distribution of injured drivers by injury type

Injury type Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Cerebral hemorrhage 55 15 70 13 4 17

11.7% 12.9% 11.9% 23.6% 26.7% 24.3%

Fracture/crushing/dislocation 230 64 294 13 8 21

48.9% 55.2% 50.2% 5.7% 12.5% 7.1%

Sprain 111 18 129 – – –

23.6% 15.5% 22.0% – – –

Rupture 21 4 25 2 0 2

4.5% 3.4% 4.3% 9.5% 0.0% 8.0%

Contusion/laceration/abrasion 34 10 44 – – –

7.2% 8.6% 7.5% – – –

Complex 19 5 24 12 5 17

4.0% 4.3% 4.1% 63.2% 100.0% 70.8%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 4.093, p = 0.536 χ² = 1.860, p = 0.602

Injury type	Injury (fatal and nonfatal)	Death
Cerebral hemorrhage	55	15	70	13	4	17
	11.7%	12.9%	11.9%	23.6%	26.7%	24.3%
Fracture/crushing/dislocation	230	64	294	13	8	21
	48.9%	55.2%	50.2%	5.7%	12.5%	7.1%
Sprain	111	18	129	–	–	–
	23.6%	15.5%	22.0%	–	–	–
Rupture	21	4	25	2	0	2
	4.5%	3.4%	4.3%	9.5%	0.0%	8.0%
Contusion/laceration/abrasion	34	10	44	–	–	–
	7.2%	8.6%	7.5%	–	–	–
Complex	19	5	24	12	5	17
	4.0%	4.3%	4.1%	63.2%	100.0%	70.8%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 4.093, p = 0.536	χ² = 1.860, p = 0.602

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by injury type (χ² = 1.860, p = 0.602). The mortality rate of the injured drivers who were dead by cerebral hemorrhage was 24.3%, followed by rupture (8.0%). Especially, the mortality rate of the injured drivers who died of complex injury type was as high as 70.8%. Except for those who injured by rupture, the mortality rates of those aged ≥65 years were much higher than the mortality rates of those aged <65 years.

3.2 Characteristics of the injured drivers in view of crash-related factors

3.2.1 Analysis of the injured drivers by day of the week

Table 8 reveals the distribution of 586 injured drivers by day of the week when the crash occurred. Among 586 casualties, 385 (65.7%) injured on the weekdays, and 201 (34.3%) injured on the weekend. There was no difference between the age groups by day of the week (χ² = 0.153, p = 0.392).

Table 8
Distributions of the injured drivers by day of the week

Day of the week Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Weekdays 307 78 385 21 10 31

65.3% 67.2% 65.7% 6.8% 12.8% 8.1%

Weekend 163 38 201 19 7 26

34.7% 32.8% 34.3% 11.7% 18.4% 12.9%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 0.153, p = 0.392 χ² = 0.192, p = 0.443

Day of the week	Injury (fatal and nonfatal)	Death
Weekdays	307	78	385	21	10	31
	65.3%	67.2%	65.7%	6.8%	12.8%	8.1%
Weekend	163	38	201	19	7	26
	34.7%	32.8%	34.3%	11.7%	18.4%	12.9%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 0.153, p = 0.392	χ² = 0.192, p = 0.443

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by day of the week (χ² = 0.192, p = 0.443). The mortality rate of the injured drivers who injured on weekdays was 8.1% (31/385) while the mortality rate of the injured drivers injured on the weekend was 12.9%. Irrespective of the day of the week, the mortality rates of those aged ≥65 years were higher than the mortality rates of those aged <65 years.

3.2.2 Analysis of the injured drivers by the time of the day

Table 9 shows the distribution of 586 injured drivers by the time of the day when the crash occurred. Among 586 casualties, 365 (62.3%) injured at night and 221 (37.7%) injured in the daytime. There was a statistically significant difference between the age groups by time of the day (χ² = 3.116, p = 0.049). For those aged <65 years, 64.0% injured at night and 36.0% injured in the daytime, which implied that about 2/3 of the crash occurred at night, but 55.2% of those aged ≥65 years injured at night and 44.8% injured in the daytime.

Table 9
Distributions of injured drivers by time of the day

Time of day Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Daytime 169 52 221 17 8 25

36.0% 44.8% 37.7% 10.1% 15.4% 11.3%

Night 301 64 365 23 9 32

64.0% 55.2% 62.3% 7.6% 14.1% 8.8%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 3.116, p = 0.049 χ² = 0.101, p = 0.751

Time of day	Injury (fatal and nonfatal)	Death
Daytime	169	52	221	17	8	25
	36.0%	44.8%	37.7%	10.1%	15.4%	11.3%
Night	301	64	365	23	9	32
	64.0%	55.2%	62.3%	7.6%	14.1%	8.8%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 3.116, p = 0.049	χ² = 0.101, p = 0.751

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by time of the day (χ² = 0.101, p = 0.751). The mortality in the daytime was 11.3%, while the mortality rate at night was 8.8%. Irrespective of the time of day, the mortality rates of those aged ≥65 years were higher than the mortality rates of those aged <65 years.

3.2.3 Analysis of injured drivers by weather condition

Table 10 gave the distribution of 586 injured drivers by weather conditions when the crash occurred. Among 586 casualties, 109 (18.6%) injured when it rains or snows and 477 (81.4%) injured when it is fair or cloudy. There was a statistically significant difference between the age groups by weather conditions (χ² = 11.229, p < 0.001). For the group aged <65 years, 21.3% injured when it rains or snows, but for those aged ≥65 years, only 7.8% injured when it rains or snows.

Table 10
Distributions of injured drivers by weather condition

Weather Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Rain/snow 100 9 109 9 1 10

21.3% 7.8% 18.6% 9.0% 11.1% 9.2%

Otherwise 370 107 477 31 16 47

78.7% 92.2% 81.4% 8.4% 15.0% 9.9%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 11.229, p < 0.001 χ² = 2.277, p = 0.127

Weather	Injury (fatal and nonfatal)	Death
Rain/snow	100	9	109	9	1	10
	21.3%	7.8%	18.6%	9.0%	11.1%	9.2%
Otherwise	370	107	477	31	16	47
	78.7%	92.2%	81.4%	8.4%	15.0%	9.9%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 11.229, p < 0.001	χ² = 2.277, p = 0.127

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups by weather conditions (χ² = 2.277, p = 0.127). The mortality rate of the injured drivers when it rains or snows was 9.2% (10/109) while the mortality rate of the injured drivers injured when it is fair or cloudy was 9.9%. Irrespective of the weather condition, the mortality rates of those aged ≥65 years were much higher than the mortality rates of those aged <65 years.

3.2.4 Analysis of injured drivers by road type

Table 11 presents the distribution of 586 injured drivers by road type where the crash occurred. Among 586 casualties, 69 (11.8%) injured on an expressway and 517 (88.2%) injured on a rural road or urban street. There was a statistically significant difference between the age groups by road type (χ² = 3.313, p = 0.043). For those aged <65 years, 13.0% injured on an expressway, but for those aged ≥65 years, only 6.9% injured on an expressway.

Table 11
Distributions of injured drivers by road type

Road type Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Expressway 61 8 69 8 0 8

13.0% 6.9% 11.8% 13.1% 0.0% 11.6%

Rural road 409 108 517 32 17 49

87.0% 93.1% 88.2% 7.8% 15.7% 9.5%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 3.313, p = 0.043 χ² = 3.955, p = 0.047

Road type	Injury (fatal and nonfatal)	Death
Expressway	61	8	69	8	0	8
	13.0%	6.9%	11.8%	13.1%	0.0%	11.6%
Rural road	409	108	517	32	17	49
	87.0%	93.1%	88.2%	7.8%	15.7%	9.5%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 3.313, p = 0.043	χ² = 3.955, p = 0.047

As for the dead, the distribution of 57 dead drivers shows that there was a statistically significant difference between the age groups by road type (χ² = 3.955, p = 0.047). The mortality rate on the expressway was 11.6% (8/69) higher than the mortality rate on a rural road or urban street (9.5%). In the case of drivers aged <65 years, the mortality rate on expressways was 13.1%, which was much higher than the rate of 0.0% for those aged ≥65 years. In the case of drivers aged ≥65 years, the mortality rate on a rural road or urban street was 15.7%, which was much higher than the rate of 7.8% for those aged <65 years, respectively.

3.2.5 Analysis of injured drivers by road shape

Table 12 shows the distribution of 586 injured drivers by road shape where the crash occurred. The dichotomous analysis for several road shapes is as follows. Among 586, 256 (43.7%) injured at an intersection while the rest of them (330) injured at non-intersection. 286 (48.8%) injured at the straight road while the rest of them (300) injured at non-straight road. 48 (8.2%) injured at the curved road while the rest of them (538) injured at the not curved road. Chi-square test shows that there was a statistically significant difference between the age groups whether the crash occurred on a straight road or not (χ² = 3.976, p = 0.026). For those aged ≥65 years, the percentage of the injured drivers on the straight road was as low as 40.5%, but there was no difference between the age groups by other kinds of road shape.

Table 12
Distributions of injured drivers by road shape

Location of crash Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

If intersection 197 59 256 13 7 20

41.9% 50.9% 43.7% 6.6% 11.9% 7.8%

Otherwise 273 57 330 27 10 37

58.1% 49.1% 56.3% 9.9% 17.5% 11.2%

χ² test χ² = 3.027, p = 0.051 χ² = 0.394, p = 0.369

If straight 239 47 286 17 11 28

50.9% 40.5% 48.8% 7.1% 23.4% 9.8%

Otherwise 231 69 300 23 6 29

49.1% 59.5% 51.2% 10.0% 8.7% 9.7%

χ² test χ² = 3.976, p = 0.026^* χ² = 2.354, p = 0.106

If curve 37 11 48 10 0 10

7.9% 9.5% 8.2% 27.0% 0.0% 20.8%

Otherwise 433 105 538 30 17 47

92.1% 90.5% 91.8% 6.9% 16.2% 8.7%

χ² test χ² = 0.321, p = 0.343 χ² = 5.154, p = 0.020^*

If downhill 20 5 25 2 0 2

4.3% 4.3% 4.3% 10.0% 0.0% 8.0%

Otherwise 450 111 561 38 17 55

95.7% 95.7% 95.7% 8.4% 15.3% 9.8%

χ² test χ² = 0.001, p = 0.572 χ² = 0.881, p = 0.489

If tunnel 23 3 26 2 2 4

4.9% 2.6% 4.4% 8.7% 66.7% 15.4%

Otherwise 447 113 560 38 15 53

95.1% 97.4% 95.6% 8.5% 13.3% 9.5%

χ² test χ² = 1.168, p = 0.208 χ² = 0.837, p = 0.343

Location of crash	Injury (fatal and nonfatal)	Death
If intersection	197	59	256	13	7	20
	41.9%	50.9%	43.7%	6.6%	11.9%	7.8%
Otherwise	273	57	330	27	10	37
	58.1%	49.1%	56.3%	9.9%	17.5%	11.2%
χ² test	χ² = 3.027, p = 0.051	χ² = 0.394, p = 0.369
If straight	239	47	286	17	11	28
	50.9%	40.5%	48.8%	7.1%	23.4%	9.8%
Otherwise	231	69	300	23	6	29
	49.1%	59.5%	51.2%	10.0%	8.7%	9.7%
χ² test	χ² = 3.976, p = 0.026^*	χ² = 2.354, p = 0.106
If curve	37	11	48	10	0	10
	7.9%	9.5%	8.2%	27.0%	0.0%	20.8%
Otherwise	433	105	538	30	17	47
	92.1%	90.5%	91.8%	6.9%	16.2%	8.7%
χ² test	χ² = 0.321, p = 0.343	χ² = 5.154, p = 0.020^*
If downhill	20	5	25	2	0	2
	4.3%	4.3%	4.3%	10.0%	0.0%	8.0%
Otherwise	450	111	561	38	17	55
	95.7%	95.7%	95.7%	8.4%	15.3%	9.8%
χ² test	χ² = 0.001, p = 0.572	χ² = 0.881, p = 0.489
If tunnel	23	3	26	2	2	4
	4.9%	2.6%	4.4%	8.7%	66.7%	15.4%
Otherwise	447	113	560	38	15	53
	95.1%	97.4%	95.6%	8.5%	13.3%	9.5%
χ² test	χ² = 1.168, p = 0.208	χ² = 0.837, p = 0.343

As for the dead, distribution of 57 dead drivers shows that there was a statistically significant difference between the age groups whether the crash occurred on the curved road or not (χ² = 5.154, p =0.020) but, there was no difference between the age groups by other kinds of road shape.

The mortality rate on the curved road was 20.8% (10/48), which is much higher than the rate of 8.7% on not curved road. The mortality rates on the straight road and on intersection were 9.8% and 7.8%, respectively. For those aged <65 years, the mortality rate of the injured drivers on the curved road was as high as 27.0%. Except for those who injured on curved road and downhill, the mortality rates of those aged ≥65 years were much higher than the mortality rates of those aged <65 years.

3.2.6 Analysis of injured drivers by a violation

Table 13 gives the distribution of 586 injured drivers whether or not they committed violation at the time of the crash. There was a statistically significant difference between the age groups and the violation (χ² = 6.974, p = 0.008). For those aged <65 years, 13.4% committed violation while 23.3% of those aged ≥65 years committed violation.

Table 13
Distribution of injured drivers by violation

Violation Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

With violation 63 27 90 7 7 14

13.4% 23.3% 15.4% 11.1% 25.9% 15.6%

Without violation 407 89 496 33 10 43

86.6% 76.7% 84.6% 8.1% 11.2% 8.7%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 6.974, p = 0.008 χ² = 3.609, p = 0.062

Violation	Injury (fatal and nonfatal)	Death
With violation	63	27	90	7	7	14
	13.4%	23.3%	15.4%	11.1%	25.9%	15.6%
Without violation	407	89	496	33	10	43
	86.6%	76.7%	84.6%	8.1%	11.2%	8.7%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 6.974, p = 0.008	χ² = 3.609, p = 0.062

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups and violation (χ² = 3.609, p = 0.062). The mortality rate of the injured drivers who committed violation was 15.6% (= 14/90) while the mortality rate of the injured drivers who didn’t commit violation was 8.7%. The mortality rates of those aged ≥65 years were higher than the mortality rates of those aged <65 years.

3.2.7 Analysis of injured drivers by drowsiness

Table 14 presents the distribution of 586 injured drivers whether or not they were drowsy at the time of the crash. There was no difference between the age groups and drowsiness (χ² = 0.105, p = 0.475). Among 586 casualties, 34 (5.8%) was drowsy at the time of the crash.

Table 14
Distribution of injured drivers by drowsiness

Drowsiness Injury (fatal and nonfatal) Death

<65 ≥65 Total <65 ≥65 Total

Yes 28 6 34 2 0 2

6.0% 5.2% 5.8% 7.1% 0.0% 5.9%

No 442 110 552 38 17 55

94.0% 94.8% 94.2% 8.6% 15.5% 10.0%

Total 470 116 586 40 17 57

100.0% 100.0% 100.0% 8.5% 14.7% 9.7%

χ² test χ² = 0.105, p = 0.475 χ² = 0.881, p = 0.489

Drowsiness	Injury (fatal and nonfatal)	Death
Yes	28	6	34	2	0	2
	6.0%	5.2%	5.8%	7.1%	0.0%	5.9%
No	442	110	552	38	17	55
	94.0%	94.8%	94.2%	8.6%	15.5%	10.0%
Total	470	116	586	40	17	57
	100.0%	100.0%	100.0%	8.5%	14.7%	9.7%
χ² test	χ² = 0.105, p = 0.475	χ² = 0.881, p = 0.489

As for the dead, the distribution of 57 dead drivers shows that there was no difference between the age groups and drowsiness (χ² = 0.881, p = 0.489). The mortality rate of the injured drivers who were drowsy at the time of the crash was 5.9% (= 2/34) while the mortality rate of the injured drivers who were not drowsy at the time of the crash was 10.0%. The mortality rate of those who were drowsy at the time of the crash was 7.1% for those aged <65 years, but the mortality for those aged ≥65 years was 0.0%.

4 Discussion and conclusion

In Korea, as aging progresses, the ratio of the elderly who hold a driving license is steadily increasing. As the barriers to entry for taxi drivers are low, it is increasing for retirees to get a job as a taxi driver. In this study, we analyze the characteristics of the crash of male corporate-employed taxi drivers by dividing them into two groups, one of those who are aged <65 years and the others who are aged ≥65 years. Especially the injured male taxi drivers aged ≥65 years are the main concern of the analysis.

As for the characteristics related to the crash of the taxi driver, the shorter the work experience was, the more crashes occurred. The intensive safety education for the prevention of traffic crashes is necessary when the drivers begin their careers. According to the characteristics of employment status, a majority of the injured are regular workers. However, the mortality rate of the injured drivers was higher for the temporary worker than that of a regular worker. The risk of collision of a taxi driver working part-time is twice as high [18]. In Vietnam, Iran, and Israel, taxi drivers who do not earn enough income tend to increase their income by working longer hours under the increased risk of the crash [18 , 27].

In view of the characteristics of the day of the week, the number of crashes per day occurred on weekdays was greater than on the weekends, but the mortality rate was high on the weekend. This is consistent with the findings that, among 750 injured, 259(47.9%) injured on Friday/Weekend [23]. The analysis of the time of the day showed many crashes at night rather than daytime. It is reported that, among 750 injured, 258(34.4%) injured during the night shift [23]. Age-related changes in vision among older-age people and lighting have an impact on visibility and the ability to be aware of moving objects [7]. It is recommended to consider changes from night to day shift in case of night driving because a decreased ability of the older drivers to judge the speed of moving objects and distances has an impact on night-driving [21].

When it rains or snows, the crash rate was as high as 18.6%. This is similar to the findings that, among 750 injured, 207(27.7%) are injured on the weather condition of rainy and hail, overcast, fog or mist, snowing or sleeting, and some other unfavorable weather conditions [23]. In case of bad weather, there occurred crashes that cause the vehicle to slip due to the wrong speed, so it is necessary to reduce the speed, to prevent slip accident caused by the water film phenomenon, and to secure the safety distance. For the group aged <65 years, 21.3% injured when it rains or snows but for those aged ≥65 years, only 7.8% injured when it rains or snows.

The group aged ≥65 years showed a higher mortality rate than the group aged <65 years. This is consistent with previous studies showing that older drivers have a higher severity of injury from crashes than other age groups [8 –15].

Two-thirds of the crashes due to the drivers aged <65 years occurred at night. The percentage of the crashes occurring in the daytime for drivers aged ≥65 years was higher than that of the drivers aged <65 years. At night, the drivers aged <65 years are more active than the drivers aged ≥65 years. Night driving often caused carelessness, traffic violation, or drowsiness. It is believed that the majority of taxis operated at night are corporate taxis and the need for carrying as many passengers as possible resulted in a speeding, violation and getting fatigued. It is recommended for older people to drive short distances and avoid night driving if possible because they feel fatigued quickly in long-distance driving or at night [10 , 29]. According to the characteristics of road shape where the crash occurred, half of the crashes by the drivers aged ≥65 years were related to the intersection. This is related to the previous research that showed the older drivers having difficulties driving at intersections or crossroads [15 , 31].

The crashes occurred more frequently on the rural road or urban street than on the expressway. For those aged <65 years, 13.0% of them injured on the expressway but for those aged ≥65 years, only 6.9% injured on the expressway. Mortality rates were higher on expressways. The mortality rate on the expressway was much higher for the drivers aged <65 years and the mortality rate on the rural road or urban street for drivers aged ≥65 years was twice as high as that of the drivers aged <65 years. Drivers aged 85 or older are 10.6 times more likely to die than drivers aged 25–44 years old, depending on the type of road, vehicle type, seatbelt, and alcohol [9].

15.4% of the crashes found to be due to the violation. Crashes due to violations by drivers aged 65 or older are more frequent than by drivers younger than 65. In addition, the mortality rate due to the violation was twice as high as that of non-violation. Older, more educated drivers reported engaging in more unsafe driving behaviors than younger, less educated drivers [22]. Physical aging slows down the judgment, response, and operating time of driving [32]. It was reported that those who caused crashes committed traffic violations, such as signal violations, speeding, and illegal parking at least once during the past 12 months [33].

The percentage of elderly people in taxi drivers is higher than those of cargo and bus drivers, and this resulted in the increase of traffic crashes affecting not only themselves but also other people’s lives and caused large scale damage and high mortality rates [6]. Unlike the USA, it is believed that older people are entering the taxi industry without barriers because it is easy to obtain a taxi driver’s license, and there is no retirement age. The US National Road Traffic Safety Administration (NHTSA) develops a three-step rating system for licensing renewal, assessing and managing visual, mental, and physical capabilities [34]. The elderly driver’s license renewal period may be shortened, or a vision test may be required to reduce the mortality rate of older drivers [35].

The average daily working hours of Korean taxi drivers are more than 11 hours, and they work at times of many passengers, busy commuting time or late night hours. Therefore, physical and mental fatigue is high, and the working conditions are bad [6]. The change in stressful environments of taxi driving may prevent injuries [36]. Because space and time perception and cognition abilities get limited, it is desirable to reduce the risk of collision by recognizing the limits and limiting the driving time [37]. In order to prevent crashes of the elderly, it is necessary to strengthen the safety education and to manage the period of the driving qualification tests according to the age, such as 65∼69, 70∼74. Test batteries on perception, cognition, and physical abilities for older drivers may help to identify at-risk drivers [19]. In addition, a system is necessary to manage the taxi service in an integrated manner to improve the quality of life by improving the working environment for older taxi drivers.

Policies on flexible working, occupational health, and attitude management can help to manage the aging workforce [20]. Studies on measures to support truck drivers’ working ability [38] can help to design the measures for taxi drivers. Adaptation of an adequate and acceptable driving support system for older drivers may be considered [39].

There are some limitations to this study. First, this study analyzed only the crashes of a male taxi driver working for corporations. Therefore, there may be differences in the characteristics of the crashes incurred by self-employed taxi drivers or by female drivers. Second, injury data requiring four days or more sick leave were used, and crashes with sick leave within three days or minor contact were not included in the analysis. Therefore, it may be lower than other countries’ crash rates.

Nevertheless, the results of this study provide meaningful results on the characteristics of crashes of drivers working for corporations and the characteristics of crashes of those who are under 65 years old and the other who are aged 65 or older. The findings can provide basic data for the systematic prevention and education of taxi drivers.

Conflict of interest

None to report.

Funding

This research was financially supported by Hansung University.

References

Korea Standard Classification of Occupations (KSCO); https://kssc.kostat.go.kr:8443/ksscNew_web/link.do?gubun=002; 2018 [accessed on November 2018].

Statistics Korea. The 2016 Transportation Survey. 2017.

Bureau of Labor Statistics(BLS); https://www.bls.gov/ooh/transportation-and-material-moving/taxi-drivers-and-chauffeurs.htm#tab-9; 2018 [accessed on November 2018].

Baker

, Wong

, Baron

. Professional drivers: protection needed for high-risk occupation. Am. J. Public Health. 1976;66:649–54.

Johnson

, Sorlie

, Backlund

. The impact of specific occupation on mortality in the US National Longitudinal Mortality Study. Demography. 1999;36:355–67.

The Korea Transport Institute (KOTI). Analysis and measures on aging of Taxi Drivers. 2017.

Nylén

, Favero

, Glimne

, Teär Fahnehjelm

, Eklund

. Vision, light and aging: A literature overview on older-age workers. Work. 2014;47(3):399–412.

Thompson

, Baldock

, Mathias

, et al. Older drivers in rural and urban areas: Comparisons of crash, serious injury, and fatality rates. 2010.

Hanrahan

, Layde

, Zhu

, et al. The association of driver age with traffic injury severity in Wisconsin. Traffic Inj. Prev. 2009;10(4):361–7.

10.

Awadzi

, Classen

, Hall

, et al. Predictors of injury among younger and older adults in fatal motor vehicle crashes. Accid Anal Prev. 2008;40(6):1804–10.

11.

Langford

, Koppel

. The case for and against mandatory age-based assessment of older drivers. Transp Res Part F Traffic Psychol Behav. 2006;9(5):353–62.

12.

Meuleners

, Harding

, Lee

, et al. Fragility and crash over-representation among older drivers in Western Australia. Accid Anal Prev. 2006;38(5):1006–10.

13.

Baldock

MRJ

. Self-regulation of the driving behaviour of older drivers (Doctoral dissertation). 2004.

14.

Bedard

, Guyatt

, Stones

, et al. The independent contribution of driver, crash, and vehicle characteristics to driver fatalities. Accid Anal Prev.. 2002;34(6):717–27.

15.

Ryan

, Legge

, Rosman

. Age related changes in drivers’ crash risk and crash type. Accid Anal Prev. . 1998;30(3):379–87.

16.

The Korea Road Traffic Authority(KOROAD), https://taas.koroad.or.kr/web/shp/sbm/initUnityAnalsSys.do?menuId=WEB_KMP_OVT_UAS; 2018 [accessed on September 2018].

17.

Dalziel

, Job

. Motor vehicle accidents, fatigue and optimism bias in taxi drivers. Accid Anal Prev.. 1997;29(4):489–94.

18.

, Lee

, Meuleners

, et al. Prevalence and factors associated with road traffic crash among taxi drivers in Hanoi, Vietnam. Accid Anal Prev. 2013;50:451–5.

19.

Gentzler

, Smither

. A literature review of major perceptual, cognitive, and/or physical test batteries for older drivers. Work. 2012;41(Supplement 1):5381–3.

20.

Flower

, Tipton

, Milligan

. Considerations for physical employment standards in the aging workforce. Work. 2019;63(4):509–19.

21.

Varianou-Mikellidou

, Boustras

, Dimopoulos

, Wybo

, Guldenmund

, Nicolaidou

, Anyfantis

. Occupational health and safety management in the context of an ageing workforce. Saf Sci. 2019;116:231–44.

22.

Newnam

, Mamo

, Tulu

. Exploring differences in driving behaviour across age and years of education of taxi drivers in Addis Ababa, Ethiopia. Saf Sci. 2014;68:1–5.

23.

Lam

. Environmental factors associated with crash-related mortality and injury among taxi drivers in New South Wales, Australia. Accid Anal Prev. 2004;36(5):905–8.

24.

Burns

, Wilde

. Risk taking in male taxi drivers: Relationships among personality. observational data and driver records. Pers Individ Dif. 1995;18(2):267–78.

25.

Connor

, Whitlock

, Norton

, et al. The role of driver sleepiness in car crashes: a systematic review of epidemiological studies. Accid Anal Prev. 2001;33(1):31–41.

26.

Shams

, Shojaeizadeh

, Majdzadeh

, et al. Taxi drivers’ views on risky driving behavior in Tehran: A qualitative study using a social marketing approach. Accid Anal Prev. 2011;43(3):646–51.

27.

Factor

, Mahalel

, Yair

. Inter-group differences in road-traffic crash involvement. Accid Anal Prev.. 2008;40(6):2000–7.

28.

Coughlin

, Mohyde

, D’Ambrosio

, Gilbert

. Who drives older driver decisions?. MIT AgeLab. 2004.

29.

Sterns

, Sterns

, Aizenberg

, et al. Family and friends concerned about an older driver (Final FocusGroupReport). Washington, D.C.: NHTSA. 1997.

30.

Holland

. Older drivers: a literature review. Department for Transportation (No. 25). Technical Report. 2002.

31.

Hakamies-Blomqvist

. Fatal accidents of older drivers. Accid Anal Prev. 1993;25(1):19–27.

32.

Classen

, Shechtman

, Awadzi

, et al. Traffic violations versus driving errors of older adults: Informing clinical practice. AM J Occup Ther. 2010;64(2):233–41.

33.

Huang

, Zhang

, Murphy

, et al. Attitudes and behavior of Chinese drivers regarding seatbelt use. Accid Anal Prev. 2011;43(3):889–897.

34.

Hennessy

, Janke

. Clearing a Road to Driving Fitness by Better Assessing Driving Wellness: Development of California’s Prospective Three-Tier Driving-Centered Assessment System: Summary Report (No. CAL-DMV-RSS-05-215). 2005.

35.

Grabowski

, Campbell

, Morrisey

. Elderly licensure laws and motor vehicle fatalities. JAMA.. 2004;291(23):2840–6.

36.

Wang

, Delp

. Health status, job stress and work-related injury among Los Angeles taxi drivers. Work. . 2014;49(4):705–12.

37.

Levy

. The relationship of age and state license renewal policies to driving licensure rates. Accid Anal Prev. . 1995;27(4):461–7.

38.

Staats

, Lohaus

, Christmann

, Woitschek

. Fighting against a shortage of truck drivers in logistics: Measures that employers can take to promote drivers’ work ability and health. Work. 2017;58(3):383–97.

39.

Saad

, Van Elslande

. Drivers’ safety needs, behavioural adaptations and acceptance of new driving support systems. Work. 2012;41(Supplement 1):5282–7.