Unraveling the relationship between body mass index and cardiometabolic disease, dementia, and survival in old age

Abstract

Overweight and obesity, commonly defined as a body mass index (BMI) ≥25 kg/m2, affect more than half of older adults. However, the relationship between BMI and health outcomes in old age are unclear. This PhD project aims to describe the trajectories of BMI alongside other anthropometric measures in old age. Additionally, this thesis explores the associations between short- and long-term changes in BMI with cardiometabolic diseases (CMDs), dementia, and survival, highlighting modifiable factors that could modify such associations. Longitudinal data from the Swedish National study on Aging and Care-Kungsholmen (SNAC-K) and the Screening Across the Lifespan Twin study (SALT) were used.

Study I. We assessed long-term trajectories of BMI, calf circumference (CC), and mid-arm circumference (MAC) and identified factors associated with these trajectories among older adults from SNAC-K. BMI, CC, and MAC declined significantly over the 15-year follow-up. CC and MAC, proxies of muscle mass, declined earlier and more steeply than BMI. More pronounced declines of the three anthropometrics were observed in participants aged ≥78 years of age than those <78 years of age. Moreover, cardiometabolic disorders (including diabetes, heart disease, stroke, and hypertension) accelerated the rate of these declines, whereas high educational attainment and physical activity appeared to attenuate them.

Study II. We investigated the associations of BMI and its long-term change (over 25−35 years) with CMDs (including diabetes, heart disease, and stroke) in Swedish twin individuals aged >40. Participants with high BMI (≥25 kg/m2, including overweight and obesity) had a higher risk of any CMD (hazard ratio [HR] 95% confidential interval [CI] = 1.52 [1.45−1.58]) and cardiometabolic multimorbidity (having ≥2 CMDs; 1.93 [1.76−2.13]) than those with normal BMI (20−25 kg/m2). The BMI-CMD association was independent of familial factors. A favorable lifestyle could partly mitigate the impact of high BMI on CMDs. Moreover, having high BMI only earlier (1.28 [1.02−1.59]) or later in life (1.33 [1.24−1.43]) was still associated with an increased risk of CMDs compared to having a consistently normal BMI. Those with a high BMI both in earlier and later life had the highest CMD risk (1.69 [1.55−1.85]).

Study III. We examined the association between BMI change over 6 years and subsequent incident dementia in the SNAC-K cohort. There was a U-shaped association between BMI change and dementia risk. Participants with large BMI change (>10%) had a higher dementia risk than those with stable BMI (change ≤5%) (for large BMI loss, HR [95% CI] = 2.93 [1.72−4.91]; for large BMI gain, 2.61 [1.09−5.54]). Compared to APOE ɛ4 non-carriers with stable BMI, dementia risk was higher among APOE ɛ4 carriers with a large gain (9.93 [3.49−24.6]) or loss (6.66 [2.83−14.4]) of BMI.

Study IV. We assessed the impact of mid- and late-life high BMI (≥25 kg/m2) on overall survival and chronic disease (including diabetes, heart disease, stroke, and cancer)-free survival in Swedish twins aged 60 to 79 years. Older adults with high BMI had 1.4 (95% CI 0.6−2.2) years shorter disease-free survival than those with normal BMI, though their overall survival was similar. Participants with consistently high BMI from mid- to late-life and those with high BMI in mid- but not late-life had 2.2 (95% CI 1.0−3.4) and 2.6 (95% CI 0.7−4.4) years shorter chronic disease-free survival, respectively.

Conclusions. BMI, together with CC and MAC, declines over time in older adults. High BMI in mid or late life is associated with an increased risk of CMDs, and a large change of BMI in old age is related to dementia risk. Moreover, a high BMI in mid or late-life may shorten chronic disease-free survival. Together, these findings suggest that mid or late-life high BMI and largely changes in BMI may predict adverse health outcomes in old age.