I read a BBC article a couple of years ago which showed that statistically there is very little difference between the number of accidents involving 17-19 year olds (11,946) and over 70s (10,974). The death rates were higher for over 70s (59 deaths for 70-79 year olds and 52 for the over 80s) but by far the highest accident and death rates were among the 20-24 year olds (24,007 accidents leading to 173 deaths).

The problem with statistics of course is that they never reveal the whole picture. The death rates among the older age group could be so much higher simply because older people are less resistant to heavy trauma, an accident that could result in a few bumps and bruises for a youngster could often kill an elderly person. Also there are no figures to show who was to blame for any of the above accidents. Although it is widely accepted that the majority of fatal accidents are caused by excess speed, driving too slowly can be just as dangerous especially in motorway situations and this is where older drivers are more at risk.

As regards the original question, I think the current rules are adequate but I think the assessment of an elderly driver's capabilities should be done on a yearly basis and by an outside agency. Having them done by a local GP could bring issues of familiarity into play. Assessment tests should be completely unbiased and independent in my view.

The above figures are for accidents in 2011.