Weight trajectories and obesity onset between 17 and 60 years of age, and cause-specific mortality: the Obesity and Disease Development Sweden (ODDS) pooled cohort study

Background
Longitudinal data on weight change, its timing, and the age of obesity onset in relation to cause-specific mortality are limited.

Methods
From ODDS, a nationwide pooled cohort study in Sweden, we included 258,269 men and 361,784 women with at least three weight assessments between ages 17 and 60, collected in 1963–2015. We applied linear mixed-effects models to estimate weight trajectories, age of obesity onset, and age-specific weight changes between ages 17 and 60. Outcomes were all-cause and cause-specific mortality, assessed from 5 years after the last weight assessment until death, emigration, or 31 December 2020. Associations with mortality were calculated using multivariable Cox regression models.

Findings
Over a median follow-up of 23.3 years in men and 11.7 years in women, 86,673 men and 29,076 women died. The median weight change between ages 17 and 60 was 0.42 kg/year in both sexes. A steep weight gain trajectory over this period, early obesity onset, and high weight gain between ages 17 and 29 were associated with higher all-cause mortality and with 13 of 23 specific causes of death investigated in men and 12 of 19 in women. Affected causes included cardiovascular diseases (including most subtypes), cancer (including specific types), type 2 diabetes, and digestive and genitourinary diseases. Hazard ratios (95% confidence intervals) of all-cause mortality associated with obesity onset at ages 17–29 vs. never by age 60 were 1.69 (1.60–1.79) in men and 1.71 (1.55–1.88) in women; and per 0.5 kg/year weight change at ages 17–29, 1.18 (1.17–1.19) and 1.16 (1.14–1.18), respectively. Weight gain later in adulthood generally showed weaker associations, except for cancer mortality in women, where the association was similar to that observed with earlier weight gain.

Interpretation
Weight gain in adulthood, especially in young adulthood, and obesity onset before age 30 are strong risk factors for mortality from multiple non-communicable diseases, underscoring the importance of early obesity prevention. Future studies should incorporate richer confounding data and, ideally, measures of changes in central adiposity and muscle mass.

Funding
The Swedish Research Council, Swedish Cancer Society, Crafoord Foundation, Malmö General Hospital Cancer Foundation, and the Swedish Foundation for Strategic Research.