Research Article: Age-dependent relationships between body mass index and mortality: Singapore longitudinal ageing study

Date Published: July 24, 2017

Publisher: Public Library of Science

Author(s): Tze Pin Ng, Aizhen Jin, Khuan Yew Chow, Liang Feng, Ma Shwe Zin Nyunt, Keng Bee Yap, Manlio Vinciguerra.


The relationship between body mass index (BMI) with mortality risk, in particular the BMI category associated with the lowest all-cause and CVD-and-stroke mortality and the BMI threshold for defining overweight or obesity in older persons is controversial. This study investigated the age-dependent associations of BMI categories with all-cause and cardiovascular disease (CVD) and stroke mortality.

Prospective cohort study (Singapore Longitudinal Ageing Studies) of older adults aged 55 and above, followed up from 2003 to 2011. Participants were 2605 Chinese with baseline BMI and other variables. Outcome Measurement: Mortality hazard ratios (HR) for all-cause and CVD and stroke mortality.

Overall, BMI showed a U-shaped relationship with all-cause and CVD and stroke mortality, being lowest at Normal Weight-II category (BMI 23.0–24.9 kg/m2). Most evidently among the middle-aged (55–64 years), all-cause mortality risks relative to Normal Weight-II were elevated for underweight (30.0; HR = 4.05,p = 0.0423). Among the old (≥65 years), however, Overweight and Obese categories were not significantly associated with increased all-cause mortality (HR from 0.98 to 1.29), but Overweight-Obese was associated with increased CVD and stroke mortality (HR = 10.0, p = 0.0086).

BMI showed a U-shaped relationship with mortality. Among older persons aged 65 and above, the overweight-or-obese category of BMI was not associated with excess all-cause mortality.

Partial Text

Body mass index (BMI) is known to be associated with a continuum of health risks, with many studies showing a U-shaped relationship. [1–7] Along this continuum, the appropriate BMI cut-off point to define overweight and obesity that is associated with increased mortality risk is controversial. In the 1980’s to 2000’s, the U.S. Dietary Guidelines have variously defined overweight using BMI of 24.9 to 27.1. In 1985, the U.S. National Institutes of Health (NIH) recommended that overweight be defined using BMI 27.8 for men and 27.3 for women. In the 1990s, the definitive guide set by both the NIH and World Health Organization (WHO) was BMI over 25 for overweight and BMI over 30 for obesity.

Table 1 shows the demographic and health risk profiles of the study participants by BMI categories. Among the total of 2604 participants, 6.1% had BMI<18.5 (Underweight), 17.9% had BMI: 25.0–27.49 (Overweight Pre-obese-I), 8.6% had BMI: 27.5–29.99 (Overweight-Pre-obese-II) and 4.6% had BMI: ≥30.0 (Overweight Obese Class I, II, and III). Participants who were underweight were significantly older, and more likely to be current smokers at the start of the follow up. Our study confirms previous findings describing the U-shaped relationship of BMI with all-causes mortality. [1–7] In this population of middle-aged and older Asians, the BMI category associated with the lowest all-cause mortality was 23 to 24.99 kg/m2 (Normal Weight II). This is consistent with meta-analyses indicating that the all-cause mortality rate was lowest at 22.5 to 25.0 kg/m2. [12]   Source:


0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments