Abstract 3965: Association of Obesity and Weight-Change with Long-Term Mortality in the Elderly
Background: The epidemic of obesity in the USA raises fears of increasing death rates in the overweight. However age may be a modifying factor, and it is important to clarify the impact of weight and weight loss in the elderly.
Methods: 14-year follow-up data of the Systolic Hypertension in Elderly Program (SHEP), was used to study the differences in mortality among 4,624 men and women aged 60 years and older according to baseline body mass index (BMI, kg/m2) and change in weight from baseline to the third annual visit (± 2 pounds) performing univariate analyses and Cox regression models adjusting for demographic and clinical covariates.
Results: There were 1,990 fatal events from the start of the SHEP trial in 1985 through 2000. The total mortality rates for each BMI level among subjects ≥75 years old and <75 are shown in the figure⇓. The adjusted relative risk (RR) of mortality associated with BMI was not significant at any level of BMI using BMI 18.5<25 as the reference level. BMI below 18.5 was associated with very high mortality, primarily from non-CVD cause and ocurring soon after randomization, implying pre-existing comorbidity, adjusted RR 1.41 (0.99–2.03). Weight change (both loss or gain) was significantly associated with increased mortality, but only in the younger age group; RR 1.50 (1.27–1.77) for weight loss, and RR 1.31 (1.08–1.59) for weight gain.
Conclusions: In this cohort of elderly hypertensives increasing BMI was not associated with increased mortality. Weight change (both loss and gain) was significantly associated with increased mortality among those younger than 75 years of age but not in the older group.