Happiness does not make you live longer and stress-related misery will not shorten your life, according to a major new study.

Many people believe that unhappiness, caused by working conditions, stressful relationships or general dissatisfaction with one's lot, is likely to be life-shortening.

Some studies have in the past appeared to support that theory. But, according to researchers at Oxford University in Britain, they have generally not allowed for the unhappiness caused by poor health.

A study of just under one million women in Britain around the age of 60, published in the Lancet medical journal, found that it was true poor health made people unhappy.

It is also true that poor health made people live shorter lives. But it is not true, they said, that unhappiness was life-shortening.

"Illness makes you unhappy, but unhappiness itself doesn't make you ill," said Dr Bette Liu, now at the University of New South Wales in Australia. "We found no direct effect of unhappiness or stress on mortality, even in a 10-year study of a million women."

The paper is part of the UK Million Women study. The women, recruited through the breast screening programme between 1996 and 2001, filled in questionnaires about many aspects of their lives and health, including their happiness, and are still being followed up on today.

Co-author Richard Peto of the University of Oxford said: "The claim that this (unhappiness) is an important cause of mortality is just nonsense. Compare it with light smoking, where, if you smoke five to 10 a day, you are twice as likely to die in middle age."

Unhappiness may make people behave in an unhealthy way, such as eating or drinking too much or harming themselves. "But if you ask does it of itself have any direct effect on mortality, it doesn't," he said.

The current paper focuses on the answers of more than 700,000 women who were asked to rate their own health, happiness, stress, feelings of control and whether they felt relaxed.

Five out of six said they were generally happy - almost half (44 per cent) said they were usually happy and 39 per cent said they were happy most of the time - but 17 per cent said they were unhappy.

Random repetition of the questionnaire a year later to some of the sample showed their happiness levels changed little.

Those who said at the beginning that their health was poor were very likely also to say they were unhappy, the study finds.

During the 10 years of follow-up, more than 31,000 of the participants died (4 per cent). But the death rates among the unhappy were no higher than among those who said they were usually happy.

The researchers adjusted for factors like a woman's poor rating of her own health and also any treatment for conditions such as high blood pressure, diabetes, asthma, arthritis, depression or anxiety. They also took into account sociodemographic and lifestyle factors that are known to be potentially life-shortening, including smoking, deprivation and obesity.

Poor health and adverse lifestyle choices may shorten lives, say the authors, but after allowing for those things, "our large prospective study shows no robust evidence that happiness itself reduces cardiac, cancer or overall mortality".

Women were more likely to feel happy if they were older, less deprived, physically active, did not smoke, had a partner, belonged to a religious group or participated in social activities and had adequate sleep - but not too much.

"Many still believe that stress or unhappiness can directly cause disease, but they are simply confusing cause and effect," said Professor Peto.

THE GUARDIAN