highly significantp < 0.001
This change was highly significant ( p < 0.001 for difference across the study years), and largely occurred after 2003.
This change was highly significant ( p < 0.001 for difference across the study years), and largely occurred after 2003.
In mutually-adjusted analyses these effects were attenuated but marginally significant effects remained for gender, age and area of London. 3.3 Reporting of fatalities outside London In all three non-London cities, the level of coverage of both cyclist and motorcyclist fatalities was uniformly high across the whole of the study period, including 100% coverage of cyclist fatalities in Bristol and Cambridge (see Table 3 ).