Article Text

Download PDFPDF
Car manufacturers and global road safety: a word frequency analysis of road safety documents
  1. I Roberts1,
  2. R Wentz2,
  3. P Edwards1
  1. 1Department of Epidemiology and Population Health, London School of Hygiene & Tropical Medicine, London, UK
  2. 2Cochrane Injuries Group, London School of Hygiene & Tropical Medicine, London, UK
  1. Correspondence to:
 Professor I Roberts
 Department of Epidemiology and Population Health, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, UK; Ian.Roberts{at}lshtm.ac.uk

Abstract

Objective: The World Bank believes that the car manufacturers can make a valuable contribution to road safety in poor countries and has established the Global Road Safety Partnership (GRSP) for this purpose. However, some commentators are sceptical. The authors examined road safety policy documents to assess the extent of any bias.

Design: Word frequency analyses of road safety policy documents from the World Health Organization (WHO) and the GRSP.

Main outcome measures: The relative occurrence of key road safety terms was quantified by calculating a word prevalence ratio with 95% confidence intervals. Terms for which there was a fourfold difference in prevalence between the documents were tabulated.

Results: Compared to WHO’s World report on road traffic injury prevention, the GRSP road safety documents were substantially less likely to use the words speed, speed limits, child restraint, pedestrian, public transport, walking, and cycling, but substantially more likely to use the words school, campaign, driver training, and billboard.

Conclusions: There are important differences in emphasis in road safety policy documents prepared by WHO and the GRSP. Vigilance is needed to ensure that the road safety interventions that the car industry supports are based on sound evidence of effectiveness.

  • GRSP, Global Road Safety Partnership
  • WHO, World Health Organization
  • World Health Organization
  • bias
  • traffic accidents
  • Global Road Safety Partnership

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Every day, worldwide, around 30 000 people are seriously injured in road traffic crashes.1 Most of these are in low and middle income countries, and most of the casualties are pedestrians and cyclists. The World Bank believes that a partnership between business, non-governmental organizations, and governments in these countries can deliver substantial road safety improvements and has established the Global Road Safety Partnership (GRSP) for this purpose.2 The partnership includes the car makers General Motors, Ford, Daimler Chrysler, and Volvo, and the drinks multinationals Bacardi-Martini and United Distillers. However, concern has been raised that companies may only support initiatives that serve their commercial interests, rather than implementing evidence based interventions.3 We examined this concern by conducting word frequency analyses of road safety documents from the GRSP and the World Health Organization (WHO).

METHODS AND RESULTS

We quantified the prevalence of road safety terms in the summary report of the World report on road traffic injury prevention prepared by WHO and the World Bank4 (26 742 words excluding references) and in the combined annual reports (2003–05) of the GRSP (31 548 words).2 We compiled a glossary of road safety terms that included words for the different categories of road user (pedestrian, cyclist, passenger) and for different types of road safety intervention. We used word count software to count the occurrence of the words in the documents and the total word count. We divided the total number of occurrences of the relevant words by the total number of words in the document to give the prevalence of each word per 10 000 words. We compared the occurrence of words between the two documents by calculating a word prevalence ratio with 95% confidence intervals. We decided a priori that we would present those words for which there was a fourfold difference in prevalence between the documents. In the case of words that were not mentioned in one document, we presented them if their prevalence in the other document was greater than one in 10 000 words. Table 1 shows the results. Compared to the World report on road traffic injury prevention, the GRSP documents were substantially less likely to use the words speed, speed limits, child restraint, pedestrian, public transport, walk, walking, cycling, and cyclist, but substantially more likely to use the words school, campaign, driver training, and billboard.

Table 1

 Words and their prevalence in the WHO and GRSP reports

CONCLUSIONS

After the establishment of the GRSP, there were some concerns that car makers would be unlikely to promote initiatives that conflict with their commercial interests.3 Our analyses provide little reassurance in this respect. For example, whereas the World report emphasizes the importance of speed reduction, particularly to promote the safety of pedestrians, a recommendation that is based on strong evidence, the GRSP documents emphasize driver training and safety education campaigns, which is contrary to the available research evidence.

We used an objective method to assess differences in emphasis between documents. We have previously used word frequency analyses to identify optimal search strategies for systematic reviews.5,6 Although word frequency analysis is similar in some respects to discourse analysis, a technique that has been widely used in the analyses of policy documents, our use of word frequency analysis in this particular context is novel and our results should be seen as preliminary, warranting further detailed investigation.

There are some methodologic limitations that deserve careful consideration. First, the use of a word does not necessarily imply the promotion of a particular intervention. For example, the GRSP may have referred to driver training in the context of an intervention that has been shown to be ineffective.7 Careful reading of the GRSP documents however, does not support this interpretation. Second, the GRSP annual reports may have focused on particular projects, so that the word counts from the annual reports do not represent their work more generally. We avoided this to some extent by analyzing the content of three annual reports rather than the report for a single year; nevertheless this possibility remains open to question. We accept that annual reports may be superficial and may not provide an in-depth picture of the activities of a particular organization. We examined annual reports, rather than the GRSP road safety strategy documents,8 because we wanted to assess what the partnership actually does, rather than what it believes that it should be doing. We do not doubt the depth of road safety expertise within the partnership. The concern we sought to address is whether the GRSP would be able to persuade its commercial partners, many of which are motor manufacturers, to fund road safety initiatives that might be seen to conflict with their commercial interests. Finally, the glossary was complied by the authors and we may have included a biased selection of terms. Although the glossary was complied before the analyses were undertaken, we accept that there is still the potential for bias.

Implications for prevention

Although we welcome the contribution of the car makers in tackling the global road safety epidemic, we believe that vigilance is needed to ensure that the interventions that the industry supports are based on sound evidence of effectiveness.9

Key points

  • There were important differences in emphasis between the road safety activities of the Global Road Safety Partnership (GRSP), an organization that includes several car makers, and those advocated by the World report on road traffic injury prevention by the World Health Organization and the World Bank. Those of the GRSP were less likely to be supported by strong scientific evidence.

  • Vigilance is needed to ensure that road safety policies promoted by organizations with potential financial conflicts of interest are based on sound evidence of effectiveness.

REFERENCES

Footnotes

  • Competing interests: none.