Pergamon

Accid. Anal. and Prev., Vol. 30, No. 2, pp. 161-168, 1998 0 1998 Elsevier Science Ltd

All rights reserved. Printed in Great Britain OOOI-4575/98 $19.00 + 0.00

PII: S0001-4575(97)00089-4

PEDAL CYCLE HELMET EFFECTIVENESS: A FIELD STUDY OF PEDAL CYCLE ACCIDENTS

ANDREW MCINTOSH’ *, B. DOWDELL~ and N. SVENSSON’

‘Department of Safety Science, The University of New South Wales, Department of Safety and Science, Sydney,

N.S.W. 2052, Australia and

‘Roads and Traffic Authority, Sydney, New South Wales, Australia

(Received31 July 1995; in revisedform 8 May 1997)

Abstract-The paper describes a study of pedal cycle accidents focussing on helmet effectiveness. Only accidents in which a helmet was worn and received an impact were studied. Cases were collected from hospital accident and emergency units, through the police and via direct contact. Forty-two cases, all helmeted were examined. There were four fatal accidents; all four involved a collision with a motor vehicle. Nine cases experienced a head injury of AIS severity 22, although there were no skull fractures. Helmet impacts tended to be close to the rim anterio-laterally. The majority of non-fatal (AIS>2) head injury cases received a helmet impact to the anterio-lateral rim, which corresponds to the temporal/parietal region of the head. This site received directly only 25% of the impacts, and of these impacts, 75% produced head injuries of at least AIS=2. Soft-shell helmets tended to disintegrate on impact, and although only a single impact occurred, a helmet should remain intact to provide protection during second impacts. There was a general increase in the percentage of subjects injured or killed in accidents that involved a second vehicle compared to single vehicle accidents, 54% to 44%, respectively. This trend was stronger with cars travelling at greater than 30 km/h. 0 1998 Elsevier Science Ltd.

All rights reserved

Keywords-Pedal cyclist, Helmets, Head injury, Accidents

INTRODUCTION

Head injury is a frequent, serious and largely prevent-

able sequela of pedal cycle accidents. In the State of New South Wales (NSW), 971 pedal cyclists were killed in road accidents between January 1980 and December 1990, prior to the commencement of the study. As many non-fatal accidents remain unre- ported, these accident data understate the actual number of injured cyclists and the total number of accidents. The Australian Bureau of Statistics (1989) observed during a 1988 survey that 127,000 pedal cyclists in NSW had experienced one or more acci- dents in the preceding 12 months, and of these 22,400 (18%) sought professional medical care. Thompson et al. (1989) reported that in the U.S.A., one-third of all pedal cycle accident victims treated in emer- gency rooms and two-thirds of those admitted to hospital had head injuries. Further, in 70&80% of all fatal pedal cycle accidents, head injury was the pri- mary or contributory cause of death.

To reduce head injury helmets have been devel-

*Corresponding author. Tel: +61 2 9385 5348; Fax: +61 2 9385 5348; e-mail: a.mcintosh@UNSW.EDU.AU

oped specifically for pedal cyclists. After the introduc- tion of mandatory helmet-wearing legislation in NSW in 1991, a study of helmet wearing rates showed that 74% of children (< 16 years) and 83% of adults were wearing helmets (Institute of Transport Studies, 1993). Thompson et al. (1989) assessed the benefits of helmet usage, and concluded that the risk of head injury could be reduced by 85% and brain injury by 88% if helmets were worn. Sacks et al. (1991) claimed that through helmet use, a head injury every 4 minutes, and a fatal head injury every day could be avoided in the U.S.A. Head injury reduction was observed after the introduction of mandatory helmet- wearing legislation during 1990 in the State of Victoria, Australia (McDermott et al., 1993; Cameron et al., 1994). Although the general benefits of helmet-wearing have been studied, only a few investigations of helmet performance in real accidents have been undertaken, e.g. Williams (1989).

Investigations of helmet performance are essen- tial for the maintenance and development of good test standards. Aspects of helmet standards that require examination are: head injury tolerance criteria (pass/fail criteria) for dynamic impact tests; the

161

162 A. MCINTOSH et al.

impact velocity; the impact locations to provide ade- quate head coverage; and the impact surface.

When testing to a standard, ‘test lines’ are marked on the helmet to define the area designated for testing. Helmet areas above this line can be subjected to impact testing, whereas those areas below the line remain unassessed. In this way, the test line defines the minimum effective area of head protection. To ensure that all regions of the head have appro-

priate protection, this line must be low enough to protect the user from the majority of head impacts.

To investigate pedal cycle helmet performance

and accident conditions in a series of real cases, a field study of pedal cycle accidents was undertaken at the RTA Crashlab in Sydney. In this study, the primary selection criterion was the occurrence of a helmet impact, not the occurrence of a casualty accident. This criterion was chosen because the impact, not the injury outcome, was deemed to be the independent variable. If a subject experienced a severe impact and remained uninjured, this was of as much interest as if they were injured. It was hoped that through this method differences in helmets,

impact conditions, and consequently reasons for injury could be examined.

The specific aims were: (1) to examine the outcome of pedal cycle accidents

in which helmets were worn, with special refer- ence to head injury;

(2) to examine the types of surfaces being impacted, e.g. rigid or soft, flat or profiled;

(3) to investigate the number and location of head impacts during an accident; and

(4) to investigate relationships between head impact

conditions and injury.

METHODS

Study duration The study was conducted in two parts:

( 1) During November and December 1990, a 4-week preliminary study was undertaken. Only subjects being treated at the Prince of Wales Hospital, Sydney, were investigated.

(2) During the months January-May 1991, a l&week period, the study area was expanded to incorporate those areas detailed below. Accidents occurring within this period were investigated.

Study area All fatal accidents in the State of NSW and

selected non-fatal accidents in the Sydney Metropolitan Area were investigated. To collect sufficient cases, arrangements were made with the Police, the Forensic Medicine Division of Glebe

Coroner’s Court, and four major Sydney hospitals. The participating hospitals were the Prince of Wales, Royal Prince Alfred, Westmead hospitals, and the Royal Alexandra Hospital for Children. The study was also promoted in pedal cycle magazines, through cycling organisations and in the general media, i.e. ‘chat show’ items and the print media. This sampling

system was devised to ensure that a wide variety of cases was collected. In particular, it was hoped that both injured and uninjured cases would be collected. It was not anticipated that a collection of all cases,

or a genuine unbiased sample of all possible cases could be obtained from the public. However, all pedal cycle accidents reported through the collaborating hospitals were assessed for inclusion in the study.

Subject selection The single selection criterion for inclusion in the

study was that the cyclist sustained an impact to their helmet during an accident. If this criterion was met, but there was insufficient information regarding the case, it would then be excluded from the study. Lack

of injury or accident details, or no helmet would prevent inclusion.

Notfication procedures Fatalities. The investigators were informed

immediately of all pedal cycle fatalities. Hospitals. The accident and emergency depart-

ments and /or the trauma investigation teams of the co-operating hospitals sought informed consent from subjects that met the selection criterion. Once consent was gained, the subjects were interviewed by the investigator, their helmet borrowed for examination and their medical records, appropriate to the investi-

gation, reviewed. Direct. Subjects contacted the investigators

directly, and if they met the selection criterion, they were interviewed.

Accident investigation procedure Interview. All subjects were interviewed. In fatal

cases the police co-operated by supplying accident reports and helmets. During the interview, basic per- sonal and anthropometric data, and a description of

resultant injury were obtained. This information was both quantitative and descriptive. All information was handled in accordance with the recommendations of the New South Wales Privacy Commission. Some accident details, e.g. estimated pre-impact vehicle speeds, vehicle involvement, vehicle make and model, accident events and impacted surfaces, were also obtained. With regard to vehicle impact speed esti- mates, in many cases the cyclists had been using speedometers and were aware of their pre-collision

Pedal cycle helmet effectiveness 163

speeds. If a second vehicle was involved the existing speed limit was noted and the cyclist gave an estimate

of the vehicle’s speed. It was anticipated that investi- gation of the accident site could quantify these impact speed estimates, although this proved incorrect.

Medical records. As the study was formally approved by the research and ethics committees of all co-operating hospitals, medical records of consent- ing subjects applicable to the study were reviewed. From these records, injury details were recorded and coded according to the revised 1990 Abbreviated Injury Scale (AIS). The AIS severity is given in the text, e.g. AIS 3 denotes an AIS injury severity equal to 3.

Fatal accidents. Autopsy reports and permission to witness the autopsies were sought from the appro- priate authorities.

Helmet collection. The helmets were collected at the time of the interview or from the police.

Accident site investigation. In the majority of

cases, accident sites were visited. The accident scene was documented diagrammatically and photograph- ically to record any information that would be useful in estimating the involved vehicles’ impact velocities, the sequence of events, and the cause of the accident. The procedure followed standard motor-vehicle acci- dent investigation techniques, although in an abbrevi- ated form.

As the study progressed, it became obvious that the time lapse from accident to notification, and the minimal remaining evidence restricted the ability to

gain substantial information from accident site inves- tigation. Police records were useful in fatal cases, but

in the majority of cases, in particular single pedal cycle accidents, there was no remaining visible evi- dence to suggest that an accident had taken place. In the majority of cases, speed estimates for both cyclist and second vehicle were drawn from police reports, interviews and prevailing speed limits.

Helmet examination Helmet evaluation took two forms, firstly the

documentation of helmet damage and secondly the reproduction of the head/helmet impact dynamics (McIntosh et al., 1993)

All helmets were logged and coded with an appropriate identity number. Documentation of

helmet condition and damage was then undertaken. Helmet details, such as make, model, date of manu-

facture, conformity to standards and size, were recorded. The helmets were all photographed, in particular damaged areas. The locations of foam deformation and cracking were given a numeric code. If helmet material had separated, this was also recorded. The depth of foam deformation and area

of deformation were measured, as well as crack length. Cracking of the expanded polystyrene foam liner was identified as full-thickness or partial-thick-

ness cracking.

RESULTS

Sample size Forty-two cases were examined over a 20-week

period. The mean cyclist age at the time of accident was 27 years with a range from 7 to 5 1 years. Seventy- four percent of the subjects were male and 26% female. All the cases, except two, were from the study region. Two cases were also examined from Canberra; they had contacted the investigator in response to promotion in the general media. As the main selection criterion was the occurrence of head impact, it was

not considered that these two cases would bias the sample.

Accident details Involvement of other vehicles. Twenty-six (62%)

accidents involved another vehicle. The 26 vehicles comprised: 65% sedans, 15% utilities/vans, 4% trucks and 15% another pedal cycle. The 16 remaining cases were single cycle accidents.

Causes of accidents. In 18 cases, a second vehicle ran into the cyclist. Eight cyclists collided or clipped a second vehicle (four moving, four stationary). In 10 cases, the cyclists lost control of their bicycles. Three cases involved bicycle malfunction. In three cases, the cyclist ran into, or over, a stationary object.

Weather and road conditions. The weather was clear in 95% of the cases, and the road surface, either road or cycle way, was paved in all but one case.

Vehicle collision speed. The estimated pre-colli- sion cyclist velocities were between 0 and 60 km/h, and those of the second vehicle were between 0 and 110 km/h. The velocity was estimated from the infor- mation obtained from the police, the interview, and the accident site, as described previously. In many cases, the cyclists used speedometers. Due to the paucity of normal accident reconstruction data, e.g. skid marks or vehicle deformations, velocity estimates

are not entirely satisfactory and indicate only a general range.

Helmets Helmet types. Twenty-four different helmet

types from 11 manufacturers were collected (Table 1). Fifty-two percent of the helmets collected were hard- shell (HS), 31% soft-shell (SS) and 17% micro- shell (MS). Helmet masses were between 160 and 565 grams with a mean of 362 grams (mean helmet

164

Helmet manufacturer

A. MCINTOSH et al.

Table 1. Details of collected helmets

Model Style Standard Number collected

Bell (U.S.A.)

Giro (U.S.A.)

VI-Pro Premiere Spectrum

Image Quest

Air Attack Prolight

Britax/Safe & Sound (Australia) Davies Craig (Australia) Rosebank (Australia)

Guardian Hartop

Stackhat Gran Prix

Venturi Scott Aspen (Australia) Atom

Magnum Vortex Prolite

Apollo (Australia) Atom (Australia)

Pedla Airlite

Classic Headway (Australia) Series 301

Series 701 Brancale (Italy) XP-7 FFM (New Zealand) Vivo Vetta (Italy) Corsa-Lite

masses were for SS =222 grams, HS=468 grams and

MS = 258 grams). General condition. The general pre-accident

helmet condition was good in all cases. The mean

age of helmets, i.e. date of manufacture to accident, was 16 months, with a range from 2 to 60 months. It was only possible to establish the date of manufacture in 50% of the helmets, and therefore no trend could be established with regard to helmet age and

performance. Standards. All helmets, except one, conformed

to a set of safety standards, e.g. Australian Standard AS 2063.1,.2, the SNELL Standard (U.S.A.), or the American National Safety Standard ANSI 290.4.

Helmet impact locations. The distribution of pri- mary impacts is presented in Fig. 1. The most fre- quent site of primary impact was the anterior/lateral region of the helmet. This general region accounted for 67% of the primary helmet impacts. Minor

secondary head impacts may have occurred in approximately six cases. These were cases in which

the cyclist’s head collided with a part of a vehicle, and then the cyclist either slid off the vehicle or was thrown on to the road surface.

Impact surface. The impact surface was, in 26 cases, either the road or footpath, in nine cases a vehicle part, in three cases the windscreen, and in one case the windscreen edge and A-pillar. These surfaces, except the A-pillar, were essentially flat but with varying stiffnesses. The A-pillar would be the stiffest impact surface of those vehicle surfaces hit.

HS ss ss

MS ss

MS ss HS HS HS HS ss HS HS HS ss HS ss

HS HS MS ss ss ss

SNELL SNELL

ANSI 290.4 SNELL SNELL SNELL SNELL

AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2 AS 2063.1,.2

Unknown AS 2063.1,.2 ANSI 290.4 AS 2063.1,.2 AS 2063.1..2 AS 2063.1,.2 ANSI 290.4 AS 2063.1,.2 ANSI 290.4

3 1 2 3 1 3 1 3 1 5 1 1 2 3 1

2

1

2 1

Fig. 1. Location of primary helmet impacts. The helmet has been separated into cells and each number corresponds to the number of impacts within each cell. The number in brackets refers to the number of the nine non-fatal head injury cases that occurred within

the location,

Helmet damage. The most common helmet damage was superficial abrasion. Eighty-one percent of the helmets had more severe damage. This damage was residual foam deformation (RD) and/or half to full thickness cracking (FTC) of the liner material, with or without separation of helmet pieces (Fig. 2). A summary of helmet damage is presented in Table 2.

Ten (36%) damaged helmets exhibited material separation (excluding the crushed helmets), seven were soft-shell, two hard-shell and one micro-shell, The hard-shell helmets in this group were Rosebank Stackhats that were impacted over a thin flexible

Pedal cycle helmet effectiveness 165

(b)

Fig. 2. (a) Top: Damage to soft-shell helmet due to impact over front left rim. Cyclists impacted road surface during race. Pre- impact bicycle speed 30-40 km/h. Damage was extensive cracking and disintegration plus 25% liner deformation at impact site indi- cated by ‘X’. (b) Middle: Damage to soft-shell helmet due to lateral impact. No residual liner deformation. Cyclists collided with car. Each vehicle’s speed was between 30 and 40 km/h. (c) Bottom: Damage to hard-shell helmet showing 33% deformation of liner over right rim. Cyclist impacted road surface with pre-impact bicy-

cle speed of 20-30 km/h.

extended beam anterior to the ear. The micro-shell consisted of a sprayed on lacquer.

Retention systems. The retention systems func-

tioned in all but two cases, both fatalities. Unfortunately, it is not known whether the helmets were dislodged during the accident or if they were not fastened. Macro- and microscopic examination of the retention system revealed little, e.g. webbing elongation. In one case, a fault was observed with the webbing, although it had not contributed to any injury. In this case, stitching through a doubled over piece of webbing had failed, preventing effective retention system tightening.

Injury All but three subjects were injured, i.e. 39 (93%)

received injuries. Seventeen subjects received one or more injuries AIS 22. Eighty percent of those non- fatally injured sought treatment, either through their local General Practitioner (GP) or through Accident and Emergency (A & E), and from this group, 18% were admitted to hospital. No skull fracture occurred within the non-fatal group.

Fatal injuries. There were four fatal accidents; however, only one person died solely from head injuries. In this case, the helmet remained in place. Two people died from multiple injuries including head injuries. In one of these cases, it could not be clearly established whether the helmet was correctly worn or fastened. The other person was run over after colliding with a car. The fourth fatality suffered fatal abdominal lacerations after being hit by a car, but no head injury was sustained.

Head injury. Nine subjects suffered non-fatal head injuries, defined as AIS 2 2. AIS 1 injuries were excluded as these represent superficial, non-neurolog- ical injury. The non-fatal head injuries were all, except in one case, AIS severity 2, mostly a brief loss of consciousness for less than 5 minutes without resul- tant neurological deficit. At interview, no serious post-traumatic disability was described by any of these subjects. Some comments were made regarding a brief reduction post-trauma in short term memory and numerical skills. One person was admitted to hospital with a head injury of AIS severities 3 and 5. The single admission was not lucid on interview, and thereafter was transferred to New Zealand. In this case, police and witnesses provided valuable accident details. No skull fractures occurred in this group. Thus, a total of 10 subjects (25%), i.e. (8 x AIS=2, 1 x AIS =3 and 5, and one fatal), suffered head injuries while definitely wearing a helmet and being subjected to a direct head impact.

Other injuries. There was a total of 12 cases with at least one non-head injury of AIS 2 2; eight of these cases did not suffer a head injury. Injuries included facial, limb and pelvic fractures. Overall, the most

166 A. MCINTOSH et al.

Table 2. Nature of helmet damage

Helmet type Only liner deformation

no.: % Liner cracking + deformation

no.: % Only liner cracking

no.: % Total

no.: %

HS 3: 11 5: 18 5: 18 13: 46 ss I: 4 5: 18 4: 14 10: 36 MS 0: 0 4: 14 1: 4 5: 18 Total 4: IS 14: 50 10: 36 28: 100%

NB: This table excludes 12 cases in which there was only minimal residual deformation (~5%) and two fatal cases in which the helmets had been crushed at some stage by cars after being hislodged

frequent injury was abrasion of the limbs and/or without experiencing a head injury more severe than trunk. superficial bruising.

Accident type and injury

Single vehicle accidents and injury. Sixteen cases did not involve a collision with a second vehicle. Seven (44%) resulted in at least one injury of AIS 22, with five (31%) of these cases being head injury AIS 22. Head impacts were all against flat rigid surfaces.

Two vehicle accidents and injury. Twenty-six cases involved a collision with a second vehicle. Four of these cases were fatal and 10 resulted in at least one injury AIS 22. Four of the nine non-fatal head injuries AIS 22 resulted from a collision with a second vehicle.

Helmet type and head injury. Of the nine subjects who sustained non-fatal head injuries (AIS 22), six wore hard-shell helmets, two soft-shell and one a micro-shell helmet. This is consistent with the distri- bution of helmet types in the sample. Both the fatally injured subject and the subject with the most severe non-fatal head injury wore soft-shell helmets, the same model FFM-VIVO. It must be noted that in both cases, the probable impact velocity was very high.

Not all cyclists involved in collisions with another vehicle impacted their head against the second vehicle. In 13 of the 22 cases involving a collision with a motor vehicle it could be confirmed that the cyclists had impacted their heads against the vehicle. In nine cases impacts were against flat metal surfaces, i.e. bonnet, side or roof, three cases against windscreens and one case against the windscreen edge and A-pillar. In the other nine cases the cyclists either flew over the car or were propelled away from the vehicle.

Helmet damage and head injury. All of the nine helmets from the group non-fatal head injury group (AIS 22) were damaged. The resultant damage consisted of either liner deformation and/or liner cracking, with liner separation occurring in the soft- shell helmets.

Cycle collision speed and injury. The majority of non-fatal head injuries or other injuries of AIS 22 occurred when the cyclist was estimated to be travel- ling over 20 km/h and/or the second vehicle was travelling at greater than 30 km/h.

Helmet impact location and head injury. The primary impact locations for the group head injury AIS 2 2 were all located on the anterior/lateral aspect of the helmets. This distribution is presented in Fig. 1 in parentheses. These impacts were located in the tempero-parietal region of the head, and tended to be close to the helmet rim. In the remaining cases (AIS <2), the impact sites were distributed over a variety of locations.

DISCUSSION

The most severe non-fatal head injury occurred at high speed, unknown but possibly over 40 km/h as estimated by witnesses and the nature of the accident site (a steep hill). The single case of a fatal head injury without other injury involved a collision with a car travelling on a highway at over 100 km/h.

Helmet type and injury Only 10 of the 40 subjects that sustained a

definite helmet impact during an accident sustained a head injury, i.e. disregarding the two fatalities in which it was not certain what had occurred. Thus, 75% of the cases received a blow to their helmets

A field study of pedal cycle accidents in which helmets were worn was undertaken to examine the accident outcomes with particular reference to head injury, and the relationships between the head impact and injury. In the design and testing of helmets, it is necessary to apply realistic test conditions, so that the helmets provide protection in real accidents as opposed to laboratory tests. Some important features of helmet tests are: the number of impacts that the helmet is required to sustain; the areas of the helmet that must meet impact criteria (this prescribes the area of the head protected); and the type of material against which the helmet impacts.

In general, the study shows that 75% of the

Pedal cycle helmet effectiveness 167

possible head injury cases, i.e. all cases that received

a blunt head impact, experienced no resulting brain injury. There were four fatalities, but only one person

died from head injuries alone. All fatal accidents

involved a collision with a motor vehicle. One case

received no head injury, the person died of fatal

trunk and abdominal lacerations, and the other two

died from multiple injuries. In the single case of a

solely fatal head injury, the collision involved a car

travelling on a freeway at between 100 and 110 km/h.

The current helmet range would be unlikely to influ-

ence the outcome of these accidents. Head impacts would have been an order of magnitude far greater

than those for which the helmets are designed. A further difference between the two injury

groups was the impact site. The majority of non-fatal

head injured cases received an impact to the temporal/parietal region of the head. A single site

received directly only 25% of the impacts, and of

these impacts, 75% produced head injuries of at least

AIS = 2.

Williams ( 1989) also found that 51% of the

impacts occurring below the Test Line were in the frontal/temporal region. The significance of this

region is: first, this region is generally below the Test Line for ANSI, SNELL and AS 2063 helmet safety

standards, and therefore it does not have to meet

impact testing requirements; second, the amount of

helmet liner available to absorb and distribute impact

forces is reduced close to the rim; third, the helmet

rim is more flexible than other regions of the helmet;

and finally, the tolerance of the head to impact at

this location is probably reduced, both to direct

loading and due to angular acceleration of the head

in the coronal plane (Thibault and Gennarelli, 1989;

McIntosh et al., 1993). When these factors are considered, and coupled

with the great variability in individual tolerance to

trauma, it would appear that there is a need to

improve the protection offered by the helmets in this

region. This can be brought about most effectively

through adjustments in the positioning of the Test

Line. Whereas it would appear advantageous to

purchase a helmet in preference to one that offers

less head coverage, the extra coverage may not pro-

vide any significant improvement in impact protection as it is not required to pass an impact test.

An improved method might require that the Test

Line be measured directly from the Frankfort Plane of the headform and that the helmet provides cover- age in part down to this line. At present, it is possible

for the manufacturer to stipulate that the helmet be

positioned in such a way to pass the standard, even though this position would not be adopted by helmet

users. When the helmet is worn in another position the anterior and lateral aspects of the cranium could be exposed to direct impact. A minimum standard of

energy attenuation would be stipulated for areas of the head, which are at present insufficiently protected. Webbing attachment sites on the helmet rim should not reduce the liner thickness or result in sources of

localised loading, e.g. rivets. A potentially dangerous outcome of some

impacts was that the helmets split into a number of separate pieces. In some of these cases, the collected helmets were split without any great amount of residual deformation, the desired mode of energy attenuation being foam deformation (Gale and Mills, 1985). The cause of this is unknown, although the relatively poor tensile strength of the EPS liner com- bined with helmet/skull bending is a likely factor. In standards testing, the use of rigid head forms and uniaxial impact may hide this unwanted feature. Fortunately, in real accidents the addition of a shell

reduces this problem and in the case of micro-shells with only a slight increase in helmet mass. The shell functions not only to hold the liner together; its load distribution properties increase the amount of liner exposed during impact, and would thus reduce the magnitude of local liner bending. Although there did not appear to be any injuries caused directly by a helmet splitting into two or more pieces, the potential for injury would be increased in the event that the

cyclist was exposed to more than one head impact.

(1)

(2)

(3)

(4)

CONCLUSIONS

Forty-two pedal cycle accidents were examined in detail. In each case, a helmet impact occurred.

The average age of the subjects was 27 years, and therefore does not represent a sample including children. The majority of cases (75%) did not result in head injury. Head injury was normally low severity (concussion) without skull fracture. There were four fatalities. Fatal injuries resulted from accidents with motor vehicles. Fatalities

would be reduced most effectively through traffic engineering and education.

Pedal cycle accidents involving only the cyclists and no other vehicle are common. Impacts to the tempero-parietal region produce a greater risk of injury. It is not known whether this is a function of the head’s tolerance to impact or helmet properties. The area of head coverage provided by helmets may be insufficient. Improvements in the temporal region area may be indicated. Helmet impacts tended to be with flat, non-yielding, surfaces. There were no cases of impacts against penetrating objects.

A. MCINTOSH et al. 168

(5)

(6)

Soft-shell helmets were observed to separate into a number of pieces as a result of impact. This

would be dangerous if a second head impact occurred. Separation of the two road user groups, cyclists and motor vehicles, would reduce pedal cyclist

casualty accidents.

Acknowledgements-This project was funded by the New South Wales Roads and Traffic Authority and undertaken at the Crashlab. The authors would like to thank the Crashlab staff for their support. The generous support of the Prince of Wales Hospital, Westmead Hospital, Royal Prince Alfred Hospital and Children’s Hospital is acknowledged. Thanks are also expressed to many individuals in the Police and the Coroner’s Court.

REFERENCES

Australian Bureau of Statistics (1989) Report No. 4505.1 Bicycle Usage and Safety in NSW, May.

Cameron, M., Vulcan, A., Finch, C. and Newstead, S. (1994) Mandatory bicycle helmet use following a decade of helmet promotion in Victoria, Australia-an evaluation. Accident Analysis and Prevention 26, 3, 325-337.

Gale, A. and Mills, J. (1985) Effect of polystyrene foam

liner density on motorcycle helmet shock absorption. Plastics and Rubber Processing and Applications 5, 2, 101-108.

Institute of Transport Studies-Sydney (1993) Survey of Law Compliance and Helmet Wearing, A Draft Report, June.

McDermott, F., Lane, J., Brazenor, G. and Debney, E. (1993) The effectiveness of bicyclists helmets: a study of 1710 casualties. Journal of Trauma 34, 6, 834-844.

McIntosh, A., Kallieris, D., Mattern, R. and Miltner, E. (1993) Head and neck injury resulting from low velocity direct impact. In Proceedings of the 37th STAPP Car Crash Conference, SAE 933 112.

Sacks, J., Holmgreen, P., Smith, S. and Sosin, D. ( 1991) Bicycle-associated head injuries and deaths in the United States from 1984 through 1988. Journal of the American Medical Association 266, 21, 3016-3018.

Thibault, L. and Gennarelli, T. (1989) Brain Injury: A Research Initiative to Develop Improved Head Injury Criteria, U.S.A. Department of Transportation report HS 807 480.

Thompson, R., Pivara, F. and Thompson, D. (1989) A case-control study of the effectiveness of bicycle safety helmets. The New England Journal of Medicine 320, 21, 1361-1367.

Williams, M. (1989) The Protective Performance of Bicy- clists Helmets in Accidents. Technisearch Ltd, Royal Melbourne Institute of Technology.