The researchers say official forecasts underestimate how long people will live in the future, and therefore don't adequately anticipate the need for additional investments in health and social services and pensions for the elderly.

The new study, published in the Lancet, also predicts that regional inequality in life expectancy will increase, highlighting a need to help deprived districts catch up with affluent areas.

Researchers at Imperial College London developed statistical models using death records, including data on age, sex, and postcode, from 1981 to 2012 to forecast life expectancy at birth for 375 districts in England and Wales.

They predict that life expectancy nationally will increase for men from 79.5 years in 2012 to 85.7 in 2030, and for women from 83.3 in 2012 to 87.6 in 2030. The longevity gap between men and women has been closing for nearly half a century and will continue to get narrower.

The forecasts for 2030 are higher than those by the Office of National Statistics, by 2.4 years for men and 1.0 year for women.

People living in the longest-living areas in 2012 - found in southern England and well-off parts of London - are expected to live seven or eight years longer than those in parts of urban northern England, such as Blackpool, Liverpool and Manchester, and South Wales - equivalent to the difference in national life expectancy between the UK and Sri Lanka or Vietnam. By 2030, the gap is projected to grow to more than eight years.

Even within London, inequality between districts is stark, with residents of Kensington and Chelsea living five to six years longer than those in Barking and Dagenham or Tower Hamlets.

Professor Majid Ezzati from the School of Public Health at Imperial College London, who led the study, said: "The bigger gains in life expectancy we predict will mean pensions will have larger payouts, and health and social services will have to serve an older population than currently planned. We also forecast rising inequalities, with bigger increases in lifespan for people in affluent areas than those in disadvantaged areas. This means wealthy people will benefit more from health and social services than poor people, and therefore should be prepared to pay its costs through higher taxes.

"The NHS plays a very important role in reducing health inequality in the UK. It's vital that it receives the investment needed to continue providing high quality care to all citizens as the population lives longer. Social and economic determinants have a huge impact on population health, so we should also be concerned about how social policies that affect the disadvantaged might also contribute to rising health inequality."

To choose the best-performing forecasting method, the Imperial team first developed five mathematical models using death records from 1981 to 2001 and tested how well they predicted data from 2002 to 2012 in a head-to-head comparison. They then used the best model for the actual forecasts.

"Our methods better reflect how longevity is changing than those currently used, and our forecasts are more accurate," said Professor Ezzati.