Endocrine Abstracts

Introduction: The effect of BMI on mortality amongst the elderly may be different from younger adults. Thus, this study aims to determine the relationship between BMI and its 5-year changes on all-cause and cause-specific mortality in the elderly.

Methods: During 1992–1994, the Blue Mountains Eye Study recruited 3654 residents aged ≥49 years in Sydney, Australia. Of these, 75% of survivors returned for follow-up at 5 years. We initially examined the relationship between baseline BMI and mortality nonparametrically using cubic spline. Cox and competing risk models were used to assess associations of baseline BMI and its 5-year changes with all-cause and cause-specific mortality.

Results: Underweight persons were more likely to be older (mean 66.9, S.D. 9.9 years) and predominantly female (74%). Conversely, obese persons were more likely to be younger (mean 63.7, S.D. 8.3 years), and to have a history of pre-existing disease (hypertension, diabetes, angina, AMI and stroke; 54.5%). Amongst subjects without pre-existing disease, the relationship between baseline BMI and all-cause mortality was U-shaped, with the underweight and obese groups being predisposed to have a greater risk of death. In particular, obesity was associated with coronary heart disease (CHD; hazard ratio (HR) 2.78, 95% CI: 1.34–5.77) and cancer (HR 1.90, 95% CI: 1.08–3.12) deaths. For subjects with pre-existing disease, however, an inverse relationship was observed, with the underweight having a lower risk of death. Five-year reductions in BMI were associated with all-cause, cancer and CHD deaths.

Conclusion: Obesity affects all-cause, CHD and cancer mortality only amongst the elderly without pre-existing disease. BMI loss was associated with major causes of mortality. Understanding the varying impact of obesity on mortality amongst the elderly with and without pre-existing disease will provide clinicians and public health policymakers with critical evidence for disease management, resource planning and allocation.