Pregnancy and risk of a traffic crash

McKillop¹ emphasizes that motor vehicle crash risks can vary depending on the weather, and that adverse weather is correlated with the Canadian winter. We agree, and this was our rationale for a secondary analysis comparing each individual’s middle trimester of pregnancy to the same months exactly a year earlier. This analysis yielded a somewhat higher relative risk associated with pregnancy (odds ratio = 1.62, 95% confidence interval 1.45–1.82, p < 0.001). Evidentially, our findings cannot be attributed to seasonal fluctuations in adverse weather.²

Buchan and collegues³ suggest that a woman’s increased propensity to seek care while pregnant might bias our analysis. We tested this idea by examining three measures of crash severity during the middle trimester and baseline interval. We found that the frequency of a higher emergency triage score was increased for crashes during pregnancy (74% v. 58%, p < 0.001). The frequency of ambulance involvement was about the same (47% v. 45%, p = 0.308) as was the frequency of admission (2.6% v. 3.2%, p = 0.409). An increased propensity to seek care is unlikely to explain the magnitude of our findings.

Each letter^1,3 also motivates points of clarification. McKillop¹ is not perfectly correct in claiming that crashes occur more often in winter — only property damage crashes increase in the winter. Crashes that cause serious injury are more common during the summer.^4,5 Buchan and colleagues³ are not perfectly correct in implying that crashes can be innocent exposures for pregnant women; instead, low-impact speeds can lead to placental separation, uterine rupture or diffuse axonal damage in the fetus. Case studies in obstetrics indicate that amniotic fluid does not perfectly protect a fetus in a crash.^6,7