Estimating the global incidence of traumatic spinal cord injury.

STUDY DESIGN: Population modelling--forecasting. OBJECTIVES: To estimate the global incidence of traumatic spinal cord injury (TSCI). SETTING: An initiative of the International Spinal Cord Society (ISCoS) Prevention Committee. METHODS: Regression techniques were used to derive regional and global estimates of TSCI incidence. Using the findings of 31 published studies, a regression model was fitted using a known number of TSCI cases as the dependent variable and the population at risk as the single independent variable. In the process of deriving TSCI incidence, an alternative TSCI model was specified in an attempt to arrive at an optimal way of estimating the global incidence of TSCI. RESULTS: The global incidence of TSCI was estimated to be 23 cases per 1,000,000 persons in 2007 (179,312 cases per annum). World Health Organization's regional results are provided. DISCUSSION: Understanding the incidence of TSCI is important for health service planning and for the determination of injury prevention priorities. In the absence of high-quality epidemiological studies of TSCI in each country, the estimation of TSCI obtained through population modelling can be used to overcome known deficits in global spinal cord injury (SCI) data. The incidence of TSCI is context specific, and an alternative regression model demonstrated how TSCI incidence estimates could be improved with additional data. The results highlight the need for data standardisation and comprehensive reporting of national level TSCI data. A step-wise approach from the collation of conventional epidemiological data through to population modelling is suggested.

The global map for traumatic spinal cord injury epidemiology: update 2011, global incidence rate.

STUDY DESIGN: Literature review. OBJECTIVES: Update the global maps for traumatic spinal cord injury (TSCI) and incorporate methods for extrapolating incidence data. SETTING: An initiative of the International Spinal Cord Society (ISCoS) Prevention Committee. METHODS: A search of Medline/Embase was performed (1959-Jun/30/2011). Enhancement of data-quality 'zones' including individual data-ranking as well as integrating regression techniques to provide a platform for continued regional and global estimates. RESULTS: A global-incident rate (2007) is estimated at 23 TSCI cases per million (179,312 cases per annum). Regional data are available from North America (40 per million), Western Europe (16 per million) and Australia (15 per million). Extrapolated regional data are available for Asia-Central (25 per million), Asia-South (21 per million), Caribbean (19 per million), Latin America, Andean (19 per million), Latin America, Central (24 per million), Latin America-Southern (25 per million), Sub-Saharan Africa-Central (29 per million), Sub-Saharan Africa-East (21 per million). DISCUSSION: It is estimated that globally in 2007, there would have been between 133 and 226 thousand incident cases of TSCI from accidents and violence. The proportion of TSCI from land transport is decreasing/stable in developed but increasing in developing countries due to trends in transport mode (transition to motorised transport), poor infrastructure and regulatory challenges. TSCIs from low falls in the elderly are increasing in developed countries with ageing populations. In some developing countries low falls, resulting in TSCI occur while carrying heavy loads on the head in young people. In developing countries high-falls feature, commonly from trees, balconies, flat roofs and construction sites. TSCI is also due to crush-injuries, diving and violence. CONCLUSION: The online global maps now inform an extrapolative statistical model, which estimates incidence for areas with insufficient TSCI data. The accuracy of this methodology will be improved through the use of prospective, standardised-data registries.

Assessing the performance of motorcyclists' impact protectors in simulated ATD knee and shoulder impacts.

OBJECTIVE: Fractures are a common injury among motorcycle riders and can have serious health implications. Impact protection (IP) has been designed to help prevent fractures, yet there are conflicting opinions as to whether this IP does in fact help prevent fractures in real-world crashes. This work aimed to (1) use simulated dummy impacts to examine whether existing types of IP could reduce the force transferred to the underlying bone to below fracture tolerance levels and (2) investigate whether current European Standard (EN 1621-1) test procedures for impact protectors designed for motorcyclists are sufficient to ensure fracture protection. METHOD: Twenty-three shoulder and 7 knee IP specimens were tested using a 23-kg impactor contacting axially along the clavicle and femur of an anthropomorphic test device (ATD) at an energy level corresponding to the fracture tolerance of these bones. Sixteen IP specimens were the same as those worn by motorcycle riders involved in crashes where injury outcome was known (knee: n = 3; shoulder: n = 13) and the IP had been previously tested to EN 1621-1. Other IP tested represented a wide range of IP available for purchase at a motorcycle accessory store. Double and triple layers of IP were also tested. Energy attenuated during the dummy impacts was compared to energy attenuated when tested to EN 1621-1. RESULTS: Of the 23 shoulder IP tested, the average percentage reduction of transferred force to the shoulder from the baseline test was 7.6 ± 4.8%. The percentage reduction of transferred force to the knee from the baseline was 43.9 ± 7.5%. The entire group of knee IP tested reduced the transferred force to the knee to below the 10-kN injury threshold for the femur. There was a positive but nonsignificant correlation between the ATD test and the EN 1621-1 impact test performance, suggesting that the European standard test method likely provides a good indication of IP performance. However, given the low correlation coefficient, the relationship between IP performance in the European standard test method and injury protection remains unclear. CONCLUSION: Though the energy attenuation test method in the European standard may be an appropriate approach, distinct differences in injury protection performance observed between knee and shoulder IP indicate that there may be a need for different performance criteria for IP designated to protect different body regions.

Alcohol consumption patterns and attitudes toward drink-drive behaviours and road safety enforcement strategies.

BACKGROUND: Alcohol contributes to approximately 30% of all serious crashes. While the majority of drivers acknowledge the risks associated with drink-driving, a significant proportion of the population continue to engage in this behaviour. Attitudes towards drink-driving as well as personal alcohol consumption patterns are likely to underpin a driver's decision to drink-drive. These associations were explored in the current study. METHODS: A large (N=2994) cross-sectional online survey of a representative sample of drivers in Australia was conducted. Participants provided information about their own alcohol consumption patterns, drink-driving behaviour as well as attitudes towards drink-driving (own and others) and enforcement strategies. RESULTS: Alcohol consumption patterns differed according to age, gender and work status. Drivers who reported drink-driving behaviour and had high risk alcohol consumption patterns were less likely to agree that drink-driving leads to increased crash risk and more likely to agree they drink and drive when they believed they could get away with it. In contrast, drivers who did not report drink-driving and had low risk consumption patterns were more likely to report that the enforcement strategies are too lenient. Binary logistic regression showed that high risk alcohol consumption patterns and agreement from drivers that they drink and drive when they believe they can get away with it had the strongest associations with drink-driving. These findings highlight the relationships between one's drinking patterns, drink-drive behaviour and attitudes towards drink-driving and drink-driving enforcement CONCLUSIONS AND IMPLICATIONS: The patterns of associations that emerged suggest that drink-driving is the expression of a broader health issue for the most "at-risk" cohort of drinkers. The decision to drink and drive may result from a need borne from an alcohol dependent lifestyle exacerbated by a social acceptability of the behaviour and positive attitudes towards one's ability to drink-drive with few adverse consequences. Therefore, the broader alcohol consumption patterns of drink-drivers needs to be considered when targeting drink-drive reductions.

Self-reported speed compliance and attitudes towards speeding in a representative sample of drivers in Australia.

BACKGROUND: Vehicle speed is a major contributor to road trauma, both in terms of increased crash risk and injury severity. In Australia, approximately one third of fatal crashes occur in speed zones of 100km/h. This proportion has remained the same, despite the reduction in the number of road fatalities over the past decade. To drive further reductions in speed-related crashes, an improved understanding of the underlying determinants of speed choice is required. METHOD: A community attitude survey designed to understand speed behaviour and attitudes towards speeding was distributed to a large (N=5179) representative sample of drivers in Australia. Participants provided information regarding their normal speed choices across four different speed zones (40, 50, 60 and 100km/h), beliefs about the risks and enforcement of speeding behaviour as well as technology to reduce speeding. RESULTS: Almost half of the sample (47%) reported exceeding the speed limit in 100km/h zones, although only a small number of these drivers (<0.5%) did so by 11km/h or more. Age and sex were related to speed limit non-compliance. Males were more likely to be classified as mid-level speeders, defined as being up to 10km/h over the limit, and excessive speeders (11+km/h over the limit). Younger drivers were also more likely to be non-compliant. When compared to compliant drivers, non-compliers perceived less risk of a serious crash, reported greater likelihood of exceeding the speed limit when they believed they would not be detected, and reported a higher level of social acceptability of speeding. Only one-third of the sample reported prior knowledge of intelligent speed assist (ISA) technology, however, once explained, the majority agreed it would be personally useful (64%). Speed non-compliers were somewhat less likely to support the usefulness of ISA than speed limit compliant drivers. CONCLUSION: These findings can be used to target appropriate interventions and road safety messages, aimed at reducing speeding behaviour. Measures designed to address perceived social acceptability of speed behaviour, the increased crash risk associated with speeding, and the threat of detection are recommended.

The relationship between Motorcycle Rider Behaviour Questionnaire scores and crashes for riders in Australia.

Motorcycle riders are over-represented in road fatalities in Australia. While riders represent 18% of the road users killed each year, motorcycle registrations constitute only 4.5% of the registered vehicle fleet. The Motorcycle Rider Behaviour Questionnaire (MRBQ) was developed with a view toward understanding behaviours likely to be associated with crash risk. These include behaviours that are either intentional (such as violations of road and speed regulations and stunts) or unintentional (such as errors relating to traffic or control of the motorcycle), as well as protective behaviours related to use of safety equipment. The dual aims of the current study were, first, to determine the appropriate structure of a modified version of the MRBQ for use in a representative sample of riders in Australia and, second, to understand which MRBQ factors are associated with crash involvement. A stratified sampling procedure was undertaken to ensure the socio-economic status of local government area, age and gender of the sample was representative of the broader population of riders in New South Wales, Australia. The sample consisted of 470 riders (males=89%). Exploratory factor analysis revealed a 29-item, five factor structure was suitable on the Australian data encompassing traffic errors, speed violations, protective gear, control errors and stunts. Overall, riders reported relatively safe behaviours, with frequent use of protective gear and infrequent aberrant behaviours. However, even though infrequent, violations of speed and errors related to control of the motorcycle increased the odds of near-crash involvement, whilst stunt behaviours were associated with increased odds of crash involvement. Interventions and countermeasures need to target these specific behaviours.

Validation of the Driver Behaviour Questionnaire in a representative sample of drivers in Australia.

The Driver Behaviour Questionnaire (DBQ) is a widely used measure of driving behaviours that may increase a driver's risk of crash involvement. However, there are several different versions of the DBQ varying in terms of number of items and factor structure. The aim of the current research was to assess the construct validity of the popular 28-item four-factor DBQ solution in a representative sample of drivers in Australia. A further aim was to test the factorial invariance of the measure across gender, age and also between fleet and non-fleet drivers using multigroup confirmatory factor analyses. Data on a range of attitudes towards road safety were collected using an online survey. A stratified sampling procedure was undertaken to ensure the age, gender and location distributions of participants were representative of the Australian population. A total of 2771 responses were obtained from fully licensed motor vehicle drivers (male: 46%). Confirmatory factor analysis supported the 28-item four-factor DBQ in the Australian sample. The DBQ was also found to be gender-invariant and strong partial measurement invariance was found for drivers aged from 26 to 64, but not for younger (17-25) or older (65-75) drivers. Modifications to the DBQ suggest how the DBQ can be improved for use in these two age groups.

Effectiveness of motorcycle protective clothing: riders' health outcomes in the six months following a crash.

BACKGROUND: Little is known about the contribution of protective clothing worn in motorcycle crashes to subsequent health-related outcomes, impairment and quality of life. METHODS: A prospective cohort of 212 adult motorcyclists were recruited following presentations to hospitals or crash repair services in a defined geographic area in Australia between June 2008 and July 2009. Data was obtained from participant interviews and medical records at baseline, then by mailed survey two and six months post-crash (n=146, 69%). The exposure factor was usage of protective clothing classified as full protection (motorcycle jacket and pants), partial protection (motorcycle jacket) and unprotected (neither). Outcomes of interest included general health status (Short Form SF-36), disability (Health Assessment Questionnaire) treatment and recovery progress, quality of life and return to work in the six months post-crash. Odds ratios (OR) were estimated for categorical outcomes using multiple logistic regression to assess differences in outcomes associated with levels of protection adjusted for potential confounders including age, sex, occupation, speed and type of impact. Non-parametric procedures were used for data that was not normally distributed. RESULTS: Compared to unprotected riders, both fully and partially protected riders had fewer days in hospital and reported less pain immediately post-crash; at two months both protection groups were less likely to have disabilities or reductions in physical function. By six months there were no significant differences in disability or physical function between groups, but both protection groups were more likely to be fully recovered and returned to pre-crash work than unprotected riders. Fully protected riders achieved better outcomes than either partially or unprotected riders on most measures. There were few significant differences between the full and partial protection groups although the latter showed greater impairment in physical health two months post-crash. CONCLUSIONS: We found strong associations between use of protective clothing and mitigation of the consequences of injury in terms of post-crash health and well-being. Given this evidence it seems likely that the use of protective clothing will confer significant benefits to riders in the event of a crash.