Lower IQ alone does not predict the chances of dying early, as previous research has often suggested, concludes a study published ahead of print in the Journal of Epidemiology and Community Health.
Other influential factors, such as risky behaviour during adolescence, mental health problems, and social circumstances as an adult cancel out the impact of low IQ, the study shows.
The findings are based on almost 44,000 men, all of whom had been born between 1949 and 1951 and were called up for compulsory military service in Sweden in 1969/70.
The conscripts underwent a comprehensive range of tests to assess logic, general intelligence, verbal ability, visuo-spatial perception, and technical and mathematical abilities, to produce an overall IQ score.
The conscripts also underwent a complete health screen for physical and mental problems and provided information on their backgrounds, which was then linked to national census returns.
Information on their subsequent progress through life was drawn from national registers recording housing status in 1990 and age and cause of death between 1970 and 2003.
When deaths were analysed by IQ alone, a clear pattern emerged: the lower the IQ, the greater was the risk of dying between the ages of 40 and 54. Those who had the lowest IQ score were over three times more likely to die early in middle age than those with the highest score.
But when other risk factors, such as childhood social circumstances, adolescent behaviour and mental health, and adult social circumstances were added to the analysis, this association all but disappeared.
Adjusting for childhood factors cut the strength of the association by 40%; taking account of social factors in adult life reduced it by 73%.