Nearly a fifth of deaths linked to hospital care occur after discharge, according to mortality figures published by the NHS Information Centre.

For the first time, the centre has linked hospital data with Office for National Statistics figures for people who have died after hospital care, up to 30, 60 and 90 days after an admission or procedure.

The figures highlight the extent of additional deaths which will be directly attributed to hospitals under a new risk adjusted mortality measure endorsed by the Department of Health and due to be published within months.

In November, following a review of mortality indicators the DH said the summary hospital-level mortality indicator, would be published for all trusts within six months.

One of its main differences from the hospital standardised mortality ratio used by Dr Foster Intelligence is to include deaths up to 30 days after discharge.

If taken from date of admission, the average monthly percentage of deaths occurring outside of hospital was 18.1 per cent after 30 days, 26 per cent after 60 days and 30.2 per cent after 90 days.

If taken from date of procedure, the average monthly percentage occurring outside of hospital was 17.6 per cent after 30 days, 26.1 per cent after 60 days and 30.6 per cent after 90 days.

The DH’s mortality indicator review last year decided that the inclusion of out of hospital deaths “gives a more complete picture of mortality associated with hospitalisation”.

It said: “This approach is consistent with the view that hospitals should be very interested in what happens to their patients in the period immediately following discharge.”

The Information Centre’s analysis of trends from April 2007 to August 2010 shows a significantly larger proportion of deaths occurring outside hospital during December. It has not yet published information for different trusts.