Grim Reaper calls earlier on poor homes

FACT: The risk of premature death increases significantly for people living in a rented home without a car. Reason: Unknown.

Assumptions that car and home ownership are merely markers of middle-class status, and that the well to do live longer, are too simplistic and no longer credible.

In the Health Variations Research Programme launched this month by the Economic and Social Research Council, such links will be rigorously put to the test for the first time. Researchers believe housing conditions and transport could either directly enhance - or harm - health.

The statistics supporting the Pounds 4 million programme back the case for an investigation of premature death. A child born to parents at the bottom of the socioeconomic scale has only half the chance of reaching adulthood of a child in the highest social class. A man in an unskilled household is three times as likely to die before the age of 65 as a professional man.

"These are widely recognised facts but for two decades inequalities in health have received little public attention," says Hilary Graham, the programme director.

"Now, with one in three of Britain's children living below the poverty line, they are back on the political agenda."

Professor Graham is seeking to answer why poorer people live shorter lives than the more affluent. Is it because of the lives they lead now or the disadvantages they faced at a younger age? Is it to do with where they live, the length of time they live in poverty, or is it more a result of their attitudes and lifestyles? Whatever the answer, she says there is clear evidence that health inequalities are widening.

Between 1970 and 1990, death rates among men in higher social classes fell more rapidly than men lower down the class ladder. Among men in social class 5, death rates are now higher than they were 20 years ago.

The cumulative impact of various risk factors could be the key. Children born into disadvantaged circumstances, for instance, are likely to have a low birth weight and to be obese in later life - two factors known to increase the risk of coronary heart disease.

Many cancers also have long natural histories that may stretch back to foetal life and infancy.

Some of the risk factors associated with these diseases such as diet, tobacco smoke, atmospheric pollution and dust, need long exposures. As a result, the risks of dying prematurely from diseases such as heart disease and chronic respiratory disease vary with the proportion of life spent in disadvantaged social circumstances.

Regional differences in mortality will also be examined, because during the 1980s the regional variations increased. Even within cities, death rates are increasingly unequal between areas.

Professor Graham concludes that effective health policy requires an understanding of social and environmental factors.