Chapter 3

Crash Data Relating to Cellular Telephone Use While Driving

 

Chapter 3 Table of Contents

3.1 Introduction

3.2 Fatal Analysis Reporting System

3.3 National Accident Sampling System

3.4 Oklahoma Crash Data

3.5 Minnesota Crash Data

3.6 Cellular Telephone Related Crash Experience in Japan

3.7 Case Studies

3.8 Discussion

 

3.1 Introduction

 

The National Highway Traffic Safety Administration uses a variety of data sources to identify emerging safety problems, monitor trends and evaluate the effectiveness of various countermeasures. Primary tools include the Fatal Analysis Reporting System (FARS) and the National Automotive Sampling System (NASS) funded by NHTSA, as well as police crash reports collected by the states. The FARS and NASS data sets make use of police crash reports as a source for information on crashes.

The FARS collects police crash reports along with other official records (such as driver records and available medical data). The NASS program employs trained investigators to document and photograph vehicle damage and scene data, and to gather additional information from interviews and medical records to enhance the data file.

Researchers must keep in mind that police crash reports are law enforcement documents. Generally, police officers are tasked with three primary duties at a crash site: tend to the injured, restore traffic flow, and issue citations for violations of the law. The identification and evaluation of specific pre-crash circumstances may be very difficult for factors such as cellular telephone use which are not in violation of the law.

With the exception of Minnesota and Oklahoma, no state includes a specific data element relating to cellular telephones on their police accident reports.

With the exception of Minnesota and Oklahoma, no state includes a specific data element relating to cellular telephones on their police crash reports. Cellular telephone usage, then, can only be identified when such information is obvious, made known to a police officer (or researcher in the case of NASS), and the information is recorded in the narrative section of the police crash report form.

Recently, many states have indicated that when a cellular telephone is reported to be in use, and when criminal charges are pending following a crash, they will now attempt to secure the telephone records. This was generally not the case prior to 1996.

FARS and NASS files began recording cellular telephone use as a possible driver-related factor in 1994.

The lack of a systematic data collection protocol generally leads to under reporting of specific factors of interest. As is shown later in this section, the simple and loosely defined data elements found on the Oklahoma form lead to skewed results in the FARS national database. This demonstrates the critical need for a focused data collection program that can address the relative risk of cellular telephone use while driving. Data searches were conducted using the currently available FARS and NASS files which began recording cellular telephone use as a possible driver-related factor in 1994. In addition, coded information from Oklahoma and Minnesota and derived narrative information from North Carolina police crash reports were reviewed (see Chapter 4).

 

Chapter 3- Table of Contents

Document Table of Contents

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

3.2 Fatal Analysis Reporting System

 

The Fatal Analysis Reporting System is a census of all motor vehicle related fatalities that occur within 30 days following a crash and which are recorded by police crash reports in the 50 United States and the District of Columbia. Approximately 40,000 deaths are recorded in this data file each year. In 1994 and 1995, a total of 36 and 40 cases, respectively, were identified that included cellular telephone use as a "possible distraction inside the vehicle".

Table 3-1, suggests that over half of these fatalities occurred in Oklahoma each year. Again, it is important to note that only Minnesota and Oklahoma include data elements relating to cellular telephone use on their police crash reports (see Figures 3-1 and 3-2). Minnesota showed no cellular telephone or CB radio related fatal crashes for 1994 or 1995. It can be assumed that the absence of data from highly populated and urbanized states such as New York can also be attributed to the limitations within the data sources.

Table 3-1: 1994&endash;1995 FARS Possible Cellular Telephone Related Fatalities

1994

1995

State

Frequency

%

Frequency

%

Arizona

1

2.8

--

--

California

3

8.3

4

10

Illinois

1

2.8

3

7.5

Indiana

1

2.8

--

--

Louisiana

--

--

2

5

Maryland

1

2.8

--

--

Missouri

--

--

2

5

New Jersey

--

--

1

2.5

North Dakota

1

2.8

--

--

Oklahoma

21

58.3

26

65

Oregon

1

2.8

--

--

Pennsylvania

1

2.8

1

2.5

Texas

4

11.1

1

2.5

Washington

1

2.8

--

--

Table 3-2: Most harmful event for cellular telephone related FARS cases*

Event

1994

1995

Object Struck

Frequency

Frequency

Other vehicle

7

5

Rail train

0

0

Pedestrian/Pedalcyclist

3

3

Animal

1

0

Ditch/culvert

1

0

Guardrail

1

0

Overturn

0

1

Fence

0

0

Tree

1

1

Post/Pole

1

2

Fixed Object

1

1

Embankment

0

0

Parked Vehicle

0

1

Other

0

0

Struck by other vehicle

1

2

* adjusted for Oklahoma - see text

The anomaly in the data set is directly attributable to the "cellular telephone installation" and "cellular telephone use" data elements which are found on Oklahoma's crash reports and which are discussed later in this report. For purposes of FARS reporting, the "telephone installed" variable was used by the FARS encoder as the indicator for a cellular telephone related crash. An examination of the actual police crash reports for Oklahoma by project staff, however, yields different results.

For 1994, only 2 crashes of the 21 shown in FARS could be verified as cellular telephone related. Among the remainder of crashes, 3 of the drivers were not considered to be at fault; in 8 crashes, use of the cellular telephone was unknown, and for an additional 8, the report indicated the phone was not in use at the time of the crash.

The 1995 Oklahoma crashes reported in FARS followed a similar pattern. Only 1 of the 26 reported crashes could definitely be attributed to cellular telephone use. For 10 others, the driver of the cellular telephone equipped vehicle was not considered to be at fault. In 13 cases, cellular telephone usage is unknown; 1 case showed no indication of phone installation or use, and in the final instance, the driver was reported to be asleep prior to the crash.

The total number of fatal crashes in Oklahoma where cellular telephones are known to have played a role as described in the report narratives includes 2 for 1994 and 1 for 1995. Only these cases are included in the discussion that follows. With regard to the Oklahoma data, there are a number of issues that should be raised. The first is that even with a simple check-off box, it is difficult to identify crashes from existing police crash reports. There has not been a focused effort to locate and describe such crashes in the U.S. to date.

There has not been a focused effort to locate and describe such [cellular telephone related] crashes in the U.S. to date.

Although the information in FARS can be misleading with regard to the proportion of cellular telephone related crashes in Oklahoma, it is believed that the cases from other states are accurately coded since the FARS analysts relied on narrative information from the police crash reports as their source for pre-crash factor identification. The cases in which cellular telephones were present, but their use is unknown, may well have included cellular telephone use as an antecedent, but the drivers were fatally injured and such usage could not be verified. For those crashes in which the driver was not considered to be "at fault" because he or she was not operating the striking vehicle, it can be argued that the lack of evasive action on the part of such drivers could, at times, be related to driver inattention associated with cellular telephone use.

When examining the 1994 FARS data, it is interesting to note that cellular telephone users were the drivers of the striking vehicles in 16 of the 17 cases. In only 1 case was the cellular telephone user the operator of the struck vehicle. As shown in Table 3-2, about half of the drivers (7) struck other motor vehicles. Three struck pedestrians and pedalcyclists, 2 of whom were in the roadway, and 1 of whom was outside the roadway. Nearly one third of these vehicles ran off the road in single vehicle collisions.

In the 1995 data, a similar pattern is evident. Thirteen of the 15 cellular telephone related drivers were in the striking vehicle. Two of the cellular telephone related cars were struck by other vehicles. One third (5 of the 15) drivers struck other vehicles on the roadway, 3 more struck pedestrians and pedalcyclists, and the final 5 hit objects on the roadside.

In FARS, up to 3 driver-related factors can be recorded for each driver in a fatal crash. Tables 3-3 and 3-4 provide comparisons of the driver related factors for cellular telephone related crashes versus all crashes reported to FARS for 1994 and 1995.

In FARS for 1994, there are 86 possible driver-related factors that are reported by police as possibly playing a role in the crash. Because of the small sample size, one must be cautious when comparing the factors cited for the 17 cellular telephone/fax related crashes to the factors for the 54,514 drivers in FARS, a comparison is provided for informational purposes only. When comparing cellular telephone related driver factors to those for all fatal crashes in 1994 (Table 3-3), it can be seen that several categories are coded more frequently. These include: inattentive, driving too fast, and failure to yield. The remaining categories are roughly equivalent, or the number of observations is very small.

A similar comparison is made for 1995 FARS in Table 3-4. There are 90 possible driver related factors included in the data system to describe the pre-crash activities of 56,155 drivers involved in fatal crashes. The cellular telephone related crashes differ from the general population by showing greater instances of inattention, erratic/reckless driving, running off the road, followed by homicide, driving on the wrong side of the road, emotional, improper passing distance, and prohibited passing. The reader is reminded that the sample sizes are very small and the comparisons must be carried out with caution.

Table 3-3: Driver Related Factors for Cellular Telephone-Related Crashes* Versus All 1994 FARS Reported Crashes

Driver-Related Factor

Cellular Telephone Related Frequency

Driver-Related Percent (%)

Driver-Related 1994 Overall FARS Percent (%)

Too Fast

5

17.9

13.8

Inattentive

7

25.0

4.2

Failure to Yield

3

10.7

6.3

Run Off Road

5

17.9

19.1

Failure to Obey

2

7.1

3.7

Erratic Driving

1

3.6

3.9

Moving Vehicle

1

3.6

0.1

Improper Turn

1

3.6

1.7

Homicide

1

3.6

2.5

Cellular Phone Use Only

1

3.6

0.0

Fax Machine Usage

1

3.6

0.0

* Only those crashes attributable to cellular phone use are included for Oklahoma

Table 3-4: Driver Related Factors for Cellular Telephone-Related* Crashes Versus All 1995 FARS Reported Crashes.

Driver-Related Factor

Cellular Telephone Related Frequency

Driver-Related Percent (%)

1995 Overall FARS Percent (%)

Too Fast

3

12.0

14.3

Inattentive

5

20.0

4.1

Run Off Road

6

24.0

19.5

Erratic/Reckless Driving

2

8.0

3.5

Improper Turn

1

4.0

1.5

Homicide

2

8.0

2.5

Cell Phone Use Only

2

8.0

0.0

Wrong Side

1

4.0

1.6

Emotional

1

4.0

0.1

Passing Distance

1

4.0

0.6

Prohibited Passing

1

4.0

0.3

* adjusted for Oklahoma

 

Chapter 3- Table of Contents

Document Table of Contents

 

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

3.3 National Automotive Sampling System

 

The National Automotive Sampling System (NASS) uses trained researchers to conduct investigations on a statistically stratified random sample of all motor vehicle crashes that occur in 24 locations across the U.S. About 5,000 crashes are investigated each year from among those reported by police in these selected primary sampling units.

The NASS data forms contain the same listing of potential driver-related factors as does FARS. The NASS researchers generally conduct interviews with crash involved drivers and vehicle occupants. This provides a greater opportunity to discern cellular telephone use as a pre-crash factor. Unless the police crash report cites cellular telephone use, the NASS researcher may not identify such use during the interview. More importantly, drivers who may be found culpable are less likely to consent to being interviewed or to admit to behaviors such as cellular telephone use as a pre-crash factor.

The 1995 NASS Crashworthiness Data System (CDS) file identified 8 relevant cellular telephone cases out of 4,555; 1 in which the driver was dialing the cellular telephone, and 7 in which the driver was talking on the cellular telephone (see Table 3.5). Although these numbers are seemingly small, when weighting factors are applied, it is estimated that these cases represent 3,837 similar crashes that occurred nationally during 1995. The final case reports were reviewed at the hard copy library. A brief summary of the crash circumstances can be found in Table 3-5.

Table 3-5:NASS Case Descriptions for Cellular Telephone-Related Crashes

 

Case 1 - The driver of the vehicle entered an intersection on a flashing red light and struck a second vehicle in the side. Both the driver and the right front passenger were engaged in conversation using a mobile telephone, permanently mounted, with a remote speaker and microphone.

Case 2 - The driver was dialing a hand-held flip phone while traveling on a divided roadway. The vehicle departed the roadway to the right and struck a utility pole and rolled on to its left side.

Case 3 - The driver was talking on a portable, hand-held cellular telephone while traveling on a rain covered roadway. The vehicle drifted onto the center median and struck a utility pole.

Case 4 - While talking on a mounted cellular telephone, the driver struck the rear of the vehicle stopped ahead of him at a traffic signal.

Case 5 - The driver was reported to be hanging up his center mounted cellular telephone when he struck the rear of the vehicle stopped ahead of him at an intersection.

Case 6 - The driver was talking on his hand-held portable phone when he struck a second vehicle in a head-on configuration. Vehicle 2 reportedly was attempting a left turn on an icy road.

Case 7 - The vehicle was traveling on a two lane, divided roadway. The driver was talking on his flip phone and had a cold drink between his legs when he struck a stopped transit bus.

Case 8 - The vehicle was traveling on a road with a gentle curve to the left. The driver was engaged in conversation using a hand held portable cellular telephone. He departed the road to the right and struck a fence.

The common factor in these cases seems to be a lack of attention. All drivers were apparently traveling in a straight line, or on a gentle curve, and were not executing difficult turning maneuvers that would have required the use of both hands. For the 5 instances in which another vehicle was struck, braking action was needed in order to avoid an obvious hazard. In Cases 2, 3 and 8, the drivers were not aware that they had traveled off the roadway. This observation corresponds with the statements found in the "Public Comments" section of this report (see Chapter 2). Drivers may become so absorbed in their conversations that they are not aware of their behavior or of the driving environment.

The common factor in these cases seems to be a lack of attention.

The NASS data cites driver inattention as a driver-related pre-crash factor in about 26% of all sampled crashes for 1995. Momentary distractions such as pushing a button on a radio would appear to have a different effect on driving behavior and ability when compared to engaging in telephone conversations that last for several minutes (and therefore several miles) of travel.

The CTIA reports that an average cellular conversation lasts 2.15 minutes. With additional time for dialing and hanging up the phone, one can assume that a driver may be occupied for perhaps 2.5 minutes on average. At 35 miles per hour (mph), the vehicle is traveling at 51.3 feet per second. At 65 mph, 95.3 feet of roadway are covered each second. At average highway speeds, from 250 to nearly 500 feet of travel are covered during the 5 seconds it takes to place a call traveling at one mile a minute. At 65 mph, about 2.7 miles of roadway are traversed during an average call (includes dialing and hanging up the phone). Even brief periods of driver distraction translate into substantial distances that might be needed for defensive driving.

Chapter 3- Table of Contents

Document Table of Contents

 

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

3.4 Oklahoma Crash Data

 

The State of Oklahoma is unique in routinely collecting specific information in their police crash reports on both cellular telephone installation and cellular telephone use in crash involved vehicles. With a population of over 3 million people, 3,350,000 registered motor vehicles and nearly 75,000 reported traffic collisions annually, the state data represents a potentially useful collection of relevant information.

The Official Oklahoma Traffic Collision Report (Figure 3-1, shown on the next two pages) includes data elements that record "telephone installed" and "telephone in use" for up to 2 vehicles in a given crash. The data is collected by the investigating officer, generally at the scene of the crash. The information is based upon both observation and interviews.

According to staff members from the Oklahoma state police training academy, officers are trained to look in crash involved vehicles to see if a cellular telephone is present. Installed car phones, and large portable units are likely to be visible, but the hand-held models that make up of current sales are less likely to be detected by casual observation. If a phone is observed, then the "telephone installed" box is checked. When a phone is observed, the driver is asked if (s)he was using the phone at the time of the crash.

If a positive response is received, then the "in use" variable is checked. It is important to note that the type of use is not defined. The driver may have been dialing, answering, or talking on the cellular telephone before or during the crash. It is also possible that the cellular telephone was used only to summon assistance post-crash. The data element was intended to identify phone use at the time of the crash, but the investigating officers do not necessarily make that distinction.

The lack of strict interpretation of data collection definition presents several problems. There is a potential for under reporting. If a cellular telephone was in use, but was not visible to the investigating officer post-crash, it will not be recorded on the police crash report as "installed". It is not likely that officers would inquire about cellular telephone use in the absence of a visible phone, so the "in-use" data element may also reflect under reporting.

In addition, culpable drivers may be less inclined to admit that they were using their cellular telephones at the time of the crash. Investigating police officers are in general agreement that witnesses are the best source of information relating to driver use of cellular telephones. Witness testimony is often not available, however, and can be unreliable.

 

Figure 3-1 Official Oklahoma Traffic Collision Report

Please click on the thumbnails for larger views of the document.

Oklahoma began recording cellular telephone use on their police crash reports in July 1992. The 1992 figures shown in Table 3-6 reflect only 6 months of data. It is interesting to note that there were nearly as many observations of telephone in-use recorded in the second half of 1992 as there were for the entire year of 1993. This may be attributable to the focus on new data elements that often occurs when they are introduced to data collectors. It is unlikely that the number of cellular telephones in cars has dropped since 1992 given the national growth. For purposes of this review, we will compare information only for those periods in which data was collected for a full year.

An analysis of the Oklahoma data shows a 26% increase in the number of cellular telephones available in crash-involved cars from 1993 to 1994. Likewise, the number of phones reported as "in use" increased by 15% during this same period. For the second half of 1992, 9.3% of identified cellular telephones were in use at the time of the crash. In 1993, about 10.4% of those cellular telephones known to be available in crash involved motor vehicles were reported to be in use at the time of the crash. In 1994, this percentage decreased to 9.5%. Thus, approximately 1 in 10 of the telephones known to be in vehicles at the time of the crash were reported to have been "in use." The reader should remember, however, the uncertainties that are introduced by the data collection methods.

 

Table 3-6 Oklahoma Vehicles Involved in Collisions

x

1992*

1993

1994

93-94 Increase

Phone in Car

968

1136

1437

26.5%

Phone in Use

90

118

136

15.3%

% in Use

9.3%

10.4%

9.5%

x

*6 months

Cellular telephone use is more common in urban areas with Tulsa and Oklahoma counties accounting for 60% of cellular telephone in use crashes in 1993 and 51% in 1994 (as cited on the police crash report). Of the 77 counties in Oklahoma, phone use crashes were reported in 32 counties in both 1993 and 1994. (see Table 3-8).

In their Annual Oklahoma Traffic Accident Facts report, the state provides an analysis of the causes of collisions similar to the driver pre-crash factors used in FARS and NASS. As can be seen in Table 3-7, contributing causes for all 217,651 police re ported crashes were compared with those for the 299 cellular telephone in use crashes for the period 1992-1994. The columns show the percentages for each group. "Driver inattention" is the most frequently identified factor among cellular telephone users. It represents 17% of the factors for cellular telephone users as compared to only 9% for all crash involved drivers. The next most frequently noted conditions are "failure to yield" and "following too closely".

 

 

Table 3-7 Contributing Causes of Oklahoma Collisions, 1992-1994

x

All

(%)

Cell. Usage

(%)

Failed to Yield

19

15

Following Too Closely

11

13

Unsafe Speed

12

6

Improper Turn

11

10

Changed Lanes Unsafely

5

6

Stopped in Traffic Lane

--

1

Failed to Stop

7

6

Unsafe Vehicle

2

1

Left of Center

2

2

Improper Overtaking

--

1

Improper Parking

2

--

Inattention

9

17

DUI

4

7

Other

16

15

Table 3-8: Oklahoma Vehicles Involved in Collisions 1993 and 1994

County

1993

1994

1993

1994

County

1993

1994

1993

1994

Adair

1

2

-

-

LeFlore

9

11

3

-

Alfalpa

5

5

1

-

Lincoln

7

15

1

-

Atoka

2

2

-

-

Logan

3

16

1

1

Beaver

6

17

-

-

Love

6

8

-

-

Beckham

13

15

2

-

McClain

7

11

1

3

Blaine

6

8

1

-

McCurtain

2

8

-

-

Bryan

4

10

-

-

Mcintosh

6

16

-

-

Caddo

3

9

3

-

Major

3

9

-

-

Canadian

20

26

2

3

Marshall

3

3

1

-

Carter

13

15

1

3

Mayes

10

10

-

-

Cherokee

4

5

-

-

Murray

4

5

1

-

Choctaw

5

1

-

-

Muskogee

23

26

2

-

Cimarron

7

3

-

-

Noble

3

13

3

-

Cleveland

55

63

3

3

Nowata

1

-

-

-

Coal

-

-

-

-

Okfuskee

1

1

-

-

Comanche

24

21

6

-

Oklahoma

320

371

38

33

Cotton

1

3

-

-

Okmulgee

4

2

1

-

Craig

6

5

1

-

Osage

12

16

1

-

Creek

19

25

-

-

Ottawa

5

4

1

-

Custer

23

14

-

-

Pawnee

4

9

1

-

Delaware

4

6

-

-

Payne

17

34

3

18

Dewey

5

9

1

-

Pittsburgh

11

21

1

-

Ellis

2

6

-

-

Pontotoc

2

12

1

-

Garfield

29

26

2

1

Pottawatomie

12

18

-

-

Garvin

7

12

1

-

Pushmataha

1

3

-

-

Grady

22

22

2

3

Roger Mills

5

3

1

-

Grant

5

7

-

-

Rogers

28

40

1

3

Greer

1

3

-

-

Seminole

1

3

1

-

Harmon

1

-

-

-

Sequoyah

2

4

1

1

Harper

3

5

-

1

Stephens

6

13

1

-

Haskell

4

-

1

1

Texas

10

9

-

-

Hughes

3

5

1

-

Tillman

3

3

1

1

Jackson

9

10

2

-

Tulsa

222

247

33

37

Jefferson

1

3

1

-

Wagoner

7

9

3

-

Johnston

3

3

-

-

Washington

8

11

1

-

Kay

22

26

1

-

Washita

6

6

1

1

Kingfisher

16

27

1

1

Woods

9

3

1

-

Kiowa

6

-

-

-

Woodward

5

13

2

-

Latimer

-

-

-

-

x

x

Totals

1136

1437

118

136

Of particular interest is the fact that the FARS data for 1994, as previously described, shows the four most common identified cellular telephone related factors as "inattention", "driving too fast", "run off road" and "failure to yield". The similarity, given the long list of possible factors, is striking.

Attempts to regulate cellular telephone use by drivers have been introduced in a number of states during the past decade (see Chapter 1). These proposed bills have generally not included provisions for the systematic data collection needed to understand and quantify related performance factors (a bill was introduced in New York State in 1994 to address such data collection - see Appendix A). The data from Oklahoma is unique insofar as it attempts to record both the installation and the use of a cellular telephone at the time of the crash. The lack of rigorous guidelines for data collection compromise the utility of the data set. The results do, however, mirror trends found in other sources of information.

The NASS and FARS files, and anecdotal observations of driver performance, are similar to the Oklahoma data in demonstrating an apparent link between cellular telephone use and driver inattention. In addition, cellular telephone use is extending beyond central city limits to more rural counties, as indicated in the Oklahoma data and reported by CTIA. The number of telephones both available and in-use during or immediately following a crash are increasing rapidly. Industry sales figures, and emergency dispatch units report similar increases.

Chapter 3- Table of Contents

Document Table of Contents

 

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

3.5 Minnesota Crash Data

 

In 1991, the State of Minnesota revised its police crash report forms (see Figure 3-2). The police officers record from 0 to 2 apparent contributing factors for drivers involved in motor vehicle collisions. Among the 32 possible factors is "driver on car phone/CB - 2 way radio." According to the Minnesota Office of Traffic Safety, there were approximately 100,000 crashes in Minnesota in 1995, with about 180,000 involved drivers. Cellular telephone/CB - 2 way radio use was not cited as a factor in any of the fatal crashes. For injury producing crashes, and for property damage only, this factor was recorded 0.1% of the time.

The Minnesota Department of Public Safety has included these statistics in their 1995 edition of Crash Facts. In previous years, the number of citations was so small that the phone/radio factor was grouped in with other "miscellaneous" factors. The pre-crash factors are often determined from interviews with involved parties. The Minnesota Office of Traffic Safety data analyst who provided these statistics compared the accuracy of the cellular telephone/radio pre-crash factor to that for self reported seat belt usage after a crash.

The NASS and FARS files, and anecdotal observations of driver performance, are similar to the Oklahoma data in demonstrating an apparent link between cellular telephone use and driver inattention.

 

Figure 3-2 State of Minnesota Police Accident Report

Please click on the thumbnails for larger views of the document.

Chapter 3- Table of Contents

Document Table of Contents

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

3.6 Cellular Telephone Related Crash Experience in Japan

 

In June 1996 the National Police Agency of Japan conducted a study in which the agency attempted to ascertain the frequency of cellular telephone use as an antecedent to a motor vehicle crash. During the month of June, 129 cellular telephone related crashes were identified. Of these, 76% involved rear end collisions, 2.3% were single vehicle crashes, 2.3% were pedestrian impacts, and 19% were categorized as "others" which would presumably include intersection and lane change related collisions. The driver related factors vary somewhat from those found in the U.S. data.

Only 16% of the drivers were conversing on the phone at the time of the crash, 32% were dialing, 5.4% were hanging up the phone, and 42% were responding to a call.

Only 16% of the drivers were conversing on the phone at the time of the crash, 32% were dialing, 5.4% were hanging up the phone, and 42% were responding to a call. The large number of crashes related to handling the telephone (32% dialing, 42% answering) may be a reflection of the fact that in Japan, 94% of telephones sold in 1995 were hand-held models compared to 73% in the U.S. It is not known if installed car telephones, and voice activated models are readily available in that market.

For the 42% of drivers who were involved in a crash as a result of responding to a call, the behaviors were described as looking aside to try to pick up the telephone, being careless in driving because of hearing the phone ring, and dropping the receiver. Of all cellular telephone related crashes, only 23 (18%) included female drivers. The majority of drivers were in the 20-29 age range. A comparison of driver factors by age group is shown in Table 3-9 (below).

The Japanese data represent the only identified attempt at comprehensive data collection of cellular telephone use by crash involved drivers by a national police department for a defined period of time. The differences in vehicle design, phone design and configuration, phone use, traffic conditions and even driving habits make it impossible to extrapolate the research results to the U.S. population. The study does show that concern for the effects of cellular telephone use while driving is of international interest and that there are a number of factors that must be considered in an analysis of the data.

Table 3.9: A Comparison of Driver Age and Causal Factor

under 19

20-29

30-39

40-49

50-59

Over 60

Speaking

9

48

14

19

5

5

Dialing

2

40

18

28

10

2

Receiving

4

44

22

24

6

0

Chapter 3- Table of Contents

Document Table of Contents

 

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

 

3.7 Case Studies

Driver behaviors and specific traffic situations that serve as antecedents to cellular telephone related motor vehicle crashes are not typically recorded on police crash reports except in Oklahoma and Minnesota as previously described. Although all states have reckless driving laws and many states prohibit careless or inattentive driving, there are no state laws that specifically limit phone use while driving. As pointed out earlier, the identification of pre-crash cellular telephone use is difficult for police officers and for researchers.

As a case in point, in a special one-time study, NASS investigators identified only 10 cellular telephone-related crashes among 60,233 police crash report narratives reviewed during April, May and June of 1996. Table 3-10 describes these crashes. Unless cellular telephones are mounted in the vehicle, there may be no physical evidence based on police crash reports alone. Even when cellular telephone presence is obvious, some drivers may not admit to pre-crash phone use for fear of being considered at-fault for the crash. The best current sources for identifying cellular telephone use are witnesses. Such observations may not, however, be recorded on a police crash report.

Table 3-10: NASS GES PAR Descriptions for Cellular Telephone Related Crashes

Case 1 - The driver of the vehicle was talking on his cellular telephone to get directions when his vehicle hit a concrete island on the left, and veered through the right lane down an embankment into a tree.

Case 2 - The driver of the vehicle took his eyes off the road while attempting to "use" a cellular telephone and the vehicle veered to the right, striking a curb.

Case 3 - The driver was talking on a cellular telephone and proceeded to turn left at an intersection as the light was turning yellow/red, and turned into the path of an on-coming vehicle.

Case 4 - The driver attempted to answer the cellular telephone and ran off the road into a tree.

Case 5 - The driver, reaching for a cellular telephone, ran a red light and struck another vehicle.

Case 6 - The driver was answering his cellular telephone, when he looked up and saw a vehicle stopped in front of him and was not able to stop in time.

Case 7 - The driver "was distracted by a cellular phone" and skidded into an intersection against a red signal and struck another vehicle.

Case 8 - The driver was talking on a cellular telephone and "not paying attention" when she rear ended another vehicle stopped for a crossing pedestrian.

Case 9 - The driver was talking on a cellular telephone and did not see a red light until it was too late to stop for an on-coming vehicle.

Case 10 - The driver was talking on a cellular telephone and made a left turn into another turning vehicle in an adjacent turning lane.

The best current sources for identifying cellular telephone use are witnesses.

In a separate effort to explore some of these issues, a pilot program was established in the Baltimore-Washington-Northern Virginia area in which police

Figure 3-3 - Posted Flyer

Please click on the thumbnail for a larger view of the document.

agencies were asked to notify the Dynamic Science, Inc. (DSI) crash investigation team when a cellular telephone related crash was identified. Several jurisdictions agreed to cooperate, and posted fliers (see Figure 3-3) to remind officers of the program. Five crashes were reported over a period of 6 months. This does not necessarily reflect the incidence of such crashes since the levels of cooperation varied by jurisdiction.


In the first case, the driver of a pick up truck in a rural area of Virginia was talking on a hand-held cellular telephone according to witnesses. His vehicle (Vehicle 1) drifted to the left, causing the vehicle next to him (Vehicle 2) to leave the roadway to the left. Vehicle 1 then struck Vehicle 2, which subsequently struck a third vehicle stopped at the crossroads. The pick-up truck driver denied using the cellular telephone, when queried by the investigating state trooper. The driver also refused to provide any additional information to the research team.


The second case was reported by the state police in a suburban Virginia area. A pick up truck was being driven while a hand-held cellular telephone was in use. The driver looked down toward the cradle, and struck a Ford sedan that was ahead of him. The trooper stated he was not authorized to release personal information or a copy of the police crash report to DSI, so no further investigation was possible.


The third crash occurred in the suburbs of Washington, D.C. A 1989 Plymouth van operated by a repair service went through a red light and was struck in the side. The van operator was talking on his hand-held, flip-phone, getting directions for his next assignment and writing these directions at the same time. There was a cellular telephone holder mounted in the van, but it was broken prior to the crash.

No hands free capabilities were available to the driver. The repair service operates 5 vans with various models of hand-held cellular telephones. Each driver uses the phone 30-40 hours per month to coordinate schedules and request directions. According to a company representative, it is not unusual for drivers to take notes while talking and operating their vehicles.

The company does not currently have rules for cellular telephone use, but the attorney present during the interview advised that guidelines will be forthcoming. While the combination of cellular telephone use and writing was likely the primary cause of this crash, it underscores the potential risks of secondary tasks associated with cellular telephone use.


The fourth incident involved a young driver of a 1992 Toyota Camry. The vehicle was traveling on a rural road at 1:00 am when the center mounted, dealer installed car phone rang. The driver attempted to answer the telephone when he drifted off the road to the right and struck a light pole.


In the fifth case, a woman was driving her minivan on a rural road near her home when she became startled because her cellular telephone rang. As she reached over to retrieve the phone from its bracket, she drifted off the road to the right, sideswiping a tree. Her child, in the right front passenger position, received fatal head injuries in the collision.


The sixth case study occurred in Arizona. In 1992, DSI investigated a crash in which a driver using a hand-held cellular telephone drifted to the left on a curve and struck a police vehicle head-on, killing the police officer.


In a seventh case, the 50 year old female driver of the 1996 minivan was travelling northbound on a 2-lane undivided roadway. Ahead of the minivan was a 15 passenger school bus that was approaching a railroad crossing. In accordance with the state motor vehicle code, the school bus driver turned on his 4-way flashers, came to a stop and opened the bus doors at the grade crossing.

The driver of the minivan looked away from the roadway to pick up her cellular telephone and, when she looked back, the bus had stopped. She collided with the rear of the bus, deploying both airbags. A 6 year old restrained right front seat passenger received critical injuries in this crash.

 

Chapter 3- Table of Contents

Document Table of Contents

 

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving

3.8 Discussion

 

The taxonomy in Table 3-11 reflects each of the particular crashes that were reported to or investigated by DSI as part of this study, as well as the 3 FARS cases, 8 NASS cases and 10 NASS/GES police crash reports presented previously. In all 28 crashes, the cellular telephone user is considered "at fault." Only 1 of the telephones is known to be a hands-free mobile unit. All others were of the hand-held type or were not specifically described on the police crash report.

The crash circumstances also fall into well defined categories. Fifteen of these crashes were attributable to drivers moving out of their designated traffic lanes. An additional 8 of the 28 cases included a collision with a stopped vehicle in the same travel lane. The remaining 5 cases occurred when the driver using the cellular telephone failed to stop for a red traffic signal.

Given the tendency for individuals to deny culpability following a crash, it is probably appropriate to consider these circumstances as useful topics upon which to focus a more detailed crash investigation research program. The patterns tend to reflect the observations shown in the Public Comment in Chapter 2. Drivers often recognize that they need to be cautious while dialing, but they tend to forget they are behind the wheel once they become engrossed in conversation.

Table 3-11 Summary of Case Studies

Source

Striking/struck

Phone Type

Cell Phone Usage

Driver Error

Object Struck

NASS 01

Striking

Hands free w/ remote speaker and microphone

talking

fail to stop

vehicle

NASS 02

Striking

Hand-held, flip phone

dialing

lane tracking

pole

NASS 03

Striking

Hand-held

talking

lane tracking

pole

NASS 04

Striking

Hand-held, mounted

talking

fail to stop

vehicle

NASS 05

Striking

Hand-held, mounted

hanging up

fail to stop

vehicle

NASS 06

Striking

Hand-held

talking

lane tracking

vehicle

NASS 07

Striking

Hand-held, flip phone

talking

fail to stop

vehicle

NASS 08

Striking

Hand-held

talking

land tracking

fence

FARS 94-1

Striking

Hand-held

answering

lane tracking

vehicle

FARS 94-2

Striking

Hand-held

talking

lane tracking

vehicle

FARS 95-1

Striking

Hand-held, mobile

talking

lane tracking

vehicle

Case Study 1

Striking

Hand-held

talking

lane tracking

vehicle

Case Study 2

Striking

Hand-held

talking

fail to stop

vehicle

Case Study 3

Struck

Hand-held, flip phone

talking

fail to stop

vehicle

Case Study 4

Striking

Hand-held, center mounted, mobile phone

answering

lane tracking

tree

Case Study 5

Striking

Hand-held,vehicle mounted

answering

lane tracking

tree

Case Study 6

Striking

Hand-held, vehicle mounted, mobile phone

talking

lane tracking

vehicle

Case Study 7

Striking

Hand-held, mobile phone

reaching

fail to stop

vehicle

PCR1

Striking

unknown

talking

lane tracking

island

PCR2

Striking

unknown

use

lane tracking

curb

PCR3

Struck

unknown

talking

fail to stop

vehicle

PCR4

Striking

unknown

answer

lane tracking

tree

PCR5

Striking

unknown

reaching

fail to stop

vehicle

PCR6

Striking

unknown

answering

fail to stop

vehicle

PCR7

Striking

unknown

distracted

fail to stop

vehicle

PCR8

Striking

unknown

talking

fail to stop

vehicle

PCR9

Striking

unknown

talking

fail to stop

vehicle

PCR10

Striking

unknown

talking

lane tracking

vehicle

Note: PCR refers to Police Crash Report (formerly PAR, Police Accident Report)

Chapter 3- Table of Contents

Document Table of Contents

 

An Investigation of the Safety Implications of Wireless Communications in Vehicles Chapter 3. Crash Data Relating to Cellular Telephone Use While Driving