EDITOR—Pilkington supports a reduction in traffic speeds to 20 mph (32 kph) on the basis of research that would be methodologically unacceptable in clinical practice.1 He implies that absolute reductions in accidents can be interpreted in terms of reductions in risk to individuals.

One hypothesis is that traffic calming schemes simply redirect traffic and accidents to other roads, with no overall reduction in risk. The Transport Research Laboratory described the findings of 72 schemes to reduce traffic speeds to ≤20 mph.2 Information on accidents on surrounding roads was available from only 40 sites, and overall no significant change occurred.

Pilkington accepts the laboratory's conclusion that there was no apparent accident migration on to surrounding roads. But this is absence of evidence that accidents migrate and not evidence of absence. The overall summary statistic masks increases in accidents on surrounding roads of up to 50% in 17 of the 40 sites. Consider basing a clinical decision on these criteria: the harm to benefit odds are only 1:1.4, and the magnitudes of harmful and beneficial effects are similar (21% (range 2-50%) v 24% (3-54%) respectively).

Pilkington drew extensively from a newspaper article that got facts wrong.3 The 30 kph (19 mph) traffic calming measures in Graz, Austria, did not reduce air pollution, and local approval rose to 68% (G Sammer, 76th annual meeting of Transportation Research Board, Washington, DC, January 1997), not “8 out of 10.”1

There is an important distinction between reductions in accidents and reductions in …