OBJECTIVE:

To examine the effects of weight change during midlife on long-term mortality risk and quality of life in old age.

DESIGN:

Prospective cohort study with a 26-y follow-up.

SUBJECTS:

Socioeconomically homogeneous sample of 1657 men (born 1919-1934) who had attended health checks during the 1960s, were healthy and professionally active in 1974, and could recall their weight at the age of 25 y.

MAIN OUTCOME MEASURES:

Total mortality 1974-2000, scales of the RAND-36 (SF-36) health survey in 91% (n=1147) of the survivors in 2000.

RESULTS:

Body weight increased from 25 y of age until midlife, but not thereafter. During the 26-y follow-up, 392 men (23.7% of the initial 1974 cohort) died. Weight at 25 y of age did not predict death, but the adjusted mortality risk was significantly increased in the highest quartile of midlife weight gain (>/=15.0 kg) compared with lower quartiles (RR 1.39, 95% CI 1.12-1.73). In 2000, multivariate analyses (adjusted for body weight at the age of 25 y and in 2000, age, smoking, alcohol and subjective health and physical fitness in 1974) showed impairment in all eight RAND-36 scales (statistically significantly in seven) with increasing weight gain in midlife.

CONCLUSION:

In this homogeneous male cohort, only the largest weight gain from 25 y of age to midlife predicted long-term mortality. Weight gain sensitively affected later health-related quality of life, and zero weight gain up to midlife was associated with the best quality of life in old age.