There are insufficient data to be able to do this. You said “the difference between the estimate and the actual reading was 1000+” so lets look at a worked example. I’m going to assume someone didn’t submit a meter reading for 5 months, I don’t know what the real situation is because you haven’t given anywhere enough detail, but lets go with this for now.
August - Actual read 1000
September - Estimated read 1100, actual read not submitted 1200
October - Estimated read 1200, actual read not submitted 1400
November - Estimated read 1300, actual read not submitted 1600
December - Estimated read 1400, actual read not submitted 1800
January - Actual read 2000
The bill for January will be for 2000 - 1400 = 600 units of usage even though all your previous estimated bills have been for only 100 units of usage. That doesn’t mean that in January you’ve suddenly used all that energy in one go. You’re just “catching up” on the previous underestimated usage.
It can’t be back-dated to show the actual previous usage, because without the actual meter readings there’s no way to know where that usage actually occurred. You’re right that it could have been avoided last year if the estimates had been more accurate. This is the problem with estimates, they can vary between wildly under and wildly over the actual usage. The only way around it is to submit a monthly meter reading and so keep on top of your actual usage.
If you look back through your previous bills, and note the date and value of all the meter readings used (whether real or estimated) you should be able to see what’s happened. If you like, post the info on here and we’ll (as in, us fellow customers) will take a look.