There seems to be some confusion over the figures. There are 2 sets of figures.
There are those released every day which show identified cases of COVID-19 and 'COVID-19' deaths. These mainly reflect COVID-19 testing of hospital patients and NHS staff.
There is a second set of statistics published every week by the Office of National Statistics [ONS] of all registered deaths in England and Wales, including those at home, hospital, care homes and elsewhere. They are published with a 2 week lag and received greater coverage yesterday because the data was for the week to April 3, the first when COVID-19 had a substantial impact. As mentioned previously, and subject to caveats, these are probably the more accurate in terms of numbers, location and region but not totally clear on the cause of death.
We must expect grim reading from the ONS figures for some further weeks, particularly as they are published with a 2 week lag. As the lock-down impacts, the number of deaths should subside towards the average but we don't know what the next set of numbers for the week to April 10 will reveal or the shape of the profile thereafter. We must hope that an extra 6,000 per week above the average for the rest of the year would be extremely pessimistic; but, actually, nobody knows.
A full analysis, including assessment of the many factors impacting on the outcome, is some time off. Meanwhile, those of us who are elderly and/or vulnerable should take all precautions possible not to join the statistics.