No abstract

T he World Health Organization classifies maternal deaths due to traffic crashes as coincidental and not related to the state of pregnancy. Others have argued that pregnancy is the root cause of such deaths, because pregnant women are more susceptible to crashes. In a linked CMAJ article, Redelmeier and colleagues 1 report intriguing findings from a Canadian population-based study showing an excess risk of serious motor vehicle crashes during the second trimester of pregnancy compared with the risk before pregnancy. The study design is novel because the researchers used data from before pregnancy to use women as their own controls. The study also makes clever use of medical billing data routinely collected in Ontario with linkage to registries that supply data on socio demographic characteristics and to the Canadian Community Health Survey for data on lifestyle factors. These robust methods address many of the flaws in previous research on the same topic. However, as with all observational studies, it is difficult to make causal inferences. In this article, we assess the associations shown in the study in light of the criteria proposed by Hill: 2 temporality, consistency, biological plausibility and evidence of a dose-response effect.Redelmeier and colleagues satisfy Hill's temporality criteria because the exposure (pregnancy) precedes the outcome (serious motor vehicle crash). Moreover, the study's findings are consistent with the findings of previous research. A population-based study from North Carolina found that the highest risk of a motor vehicle crash during pregnancy occurred at 20-31 weeks' gestation, with a marked de cline in the risk of a crash thereafter, 3 which is similar to the current study's results. A decline in the risk of a motor vehicle crash among wo men in their last trimester of pregnancy could be explained by reduced numbers of women driving during the third trimester. A population-based casecontrol study that examined the association between birth defects and crashes among pregnant women also found that crashes were highest during the second trimester of a pregnancy. 4 Examination of biological plausibility requires a discussion of the effects of normal physiologic adaptations of pregnancy that allow for the growth and survival of the fetus and ensure survival of the mother. Driving requires a high level of concentration and cognitive ability to maintain and complete a number of complex tasks. If there is any impairment in the driver's cognitive ability, there may be an increased risk of a crash. 5 The physiologic changes of pregnancy have been shown to increase fatigue and sleep deprivation in pregnant women. 6 A prospective study using self-reported questionnaires showed that sleep length began to decrease during the second trimester and quality of sleep worsened during pregnancy. 7 Maternal stress is also a common feature of pregnancy. Exposure to everyday stressors through individual, societal and familial factors may encourage a physiologic response triggering negative maternal and...