As global life expectancy rises, understanding predictors of survival in extreme old age is crucial. Body mass index (BMI) is a widely used proxy for adiposity and nutritional status. In adults, a BMI between 18.5 and 24.9 kg/m 2 is considered healthy and associated with better life expectancy; yet in nonagenarians and centenarians it remains unclear whether this BMI range is ideal due to age-related physiological changes. We examined BMI and mortality associations in 780 adults aged ≥90 years (mean age = 99.25 ± 3.3 years) from the CEPH Aging Cohort. Over six years of follow-up, 88.3% of participants died. Cox models revealed that each 5 kg/m 2 increase in BMI was associated with a 16% reduction in mortality risk (HR = 0.97 per 1 kg/m 2 increase; 95% CI: 0.95–0.99; p = 0.001), with age, as the time scale and adjusted for sex, cardiovascular disease, vulnerability indicator, neurocognitive status, and smoking status. Age-specific analyses suggested that underweight status was associated with higher mortality in the oldest age groups. The accelerated failure-time model confirmed 23% longer survival with higher BMI (time ratio 1.23; p < 0.0001). Normal and overweight individuals outlived those underweight, with no survival difference between the former groups. These findings indicate that BMI–mortality associations at very advanced ages differ from those observed in younger adults, with higher BMI among nonagenarians and centenarians serving as a marker of longer subsequent survival. Age-specific weight targets may improve longevity and resilience in this rapidly growing demographic. • We examined BMI and mortality associations in 780 nonagenarians and centenarians. • Cox model: each 5 kg/m 2 BMI increase was associated to 16% lower mortality risk. • AFT model: each 5 kg/m 2 BMI increase was associated to 23% longer survival. • Nonagenarians: no BMI effect at 90–94, but a 36% higher risk at 95–99 for underweight subjects. • Centenarians: a 35% increased risk of death was showed for underweight subjects.
Rajoua et al. (Sun,) studied this question.