I AM writing in response to your front page story ‘NHS spooked by ghost patients’ (October 8), above.

When I worked for the Health Authority back in the 1970s and ’80s, this problem was already known, and was difficult to sort out. But even now, in the days of computerised databases, it is still almost impossible to arrive at a figure for the population of any geographical area.

The approach of subtracting the city council’s estimate of the city’s population from the number of people registered with Oxford GPs is simplistic in the extreme. The city council does not have a database with all of the city’s current population on it, so their estimate is just that – a projection, presumably from the 2011 census data, which itself cannot be regarded as wholly accurate.

The number of people registered with Oxford GPs is also just an estimate of Oxford’s population. Apart from those on the register who have died, or who have moved, is it possible to correctly identify those who live strictly within the city’s boundary? Oxford GPs will have non-city patients on their registers, and there will also be some Oxford residents who are not registered with any GP (like babies, for instance). And how are students counted, on either side of the equation?

Perhaps someone could tell GPs how they are supposed to know that one of their patients has died or moved away, even with computerised databases. It is more important that they spend their time treating their patients – now there’s a thought.

LEN PORTER, London Road, Headington