THE GRAND-RAPIDS DIP REVISITED

This analysis addresses an issue that has concerned road safety author ities for some 28 years: the celebrated ''Grand Rapids Dip.'' This, mo st readers will recognise, is the below-baseline excursion, which occu rs in the blood alcohol concentration (BAC) interval of .01%-.04%, of the relative risk curve for accident risk versus blood alcohol, derive d from the 1964 Grand Rapids Study data. The present analysis has its starting place in the explanation advanced by Allsop, who noted that t he case/control comparisons were biased due to the disproportionate re presentation of demographic subgroups in different blood alcohol conce ntration class intervals. Indeed, when relative-risk curves are derive d separately for subgroups of differing drinking habits. the resulting separate risk curves all show monotonic increases at all blood alcoho l concentration ranges. Such separate relative risk curves are unpopul ar, and most of the road safety community pays them little heed. Thus, the original concept of the ''dip'' remains with us. For this reason, we have derived, using a simple but realistic statistical model, a si ngle relative-hazard curve from the Grand Rapids data, one that is fre e from the distortion introduced by unequal representation of differen t demographic subgroups in different blood alcohol concentration class intervals. This curve indicates that accident risk increases with inc reased blood alcohol concentration regardless of self-reported drinkin g frequency. However more frequent drinkers have less risk at all bloo d alcohol concentration levels, including zero, than less frequent dri nkers at the times and places sampled.