We read with interest the paper by Borrell and colleagues¹ in the June issue of Injury Prevention which reported on the relation between educational level and the risk of transportation injury mortality in nine European cohorts. An inverse association was reported in men but not women. We believe that findings from the field of cognitive (intelligence) epidemiology² may inform the discussions pertaining to the underlying mechanisms. Here, we define intelligence (denoted as IQ and assessed using written psychometric tests) as an individual’s ability to learn, reason, and solve problems.

Recent reports of extended follow ups of individuals who sat intelligence tests in childhood reveal relations between IQ scores and later health outcomes, including total mortality, coronary heart disease, and suicide. That is, people with lower IQ in early life experience a greater risk of these outcomes in adulthood.² For the most examined outcome—total mortality—the relation with early IQ is strong, incremental, and consistent across a number of study populations and research groups.² An inverse association between early adult IQ and later motor vehicle accident mortality in Australian Vietnam Army conscripts has also been found.³

Education and IQ are moderately strongly correlated; therefore, education may in part be a proxy for IQ. Some of the mechanisms that we have advanced to explain the associations between IQ and total mortality² may also have relevance to transportation injury mortality. First, IQ may be related to behaviors conducive to health which lie on the IQ-mortality pathway. For transportation injury mortality, such behaviors may include a greater use of protective devices (for example, cycle helmets, seat belts) in people with higher IQ scores. This relates to the “differential exposure” hypothesis advanced by Borrell et al.¹ Some support for this suggestion can be found in studies which show that individuals with higher educational attainment are more likely to wear cycle helmets than those with more basic qualifications.⁴ Second, information processing speed is a fundamental contributor to IQ differences, with which it is strongly correlated. An ability to process information rapidly and accurately is required in all forms of transportation, particularly those of higher speed such as car and motorbike use.

Whereas we are mindful of the maxim that “all findings can be explained by unmeasured confounding variables”, we think that intelligence differences are worth considering at least as a partial explanation for the association between educational level and transportation injury mortality.