Anthropometric measurements as predictors of mortality in a community population aged 70 years and over.

Corrected arm muscle area (CAMA), triceps skin-fold thickness (TSF) and body mass index (BMI) were measured in a community sample of 758 people who were then followed for 40-46 months. Percentile values were calculated for each sex in 5-year age groups. The relative risk of death of those in the upper and lower percentiles was compared with those between the 10th and 90th percentiles, controlling for age and sex. Subjects below the 5th percentile for CAMA, TSF and BMI and between the 5th and 10th percentile for CAMA had a significantly increased risk of death. There was no increased risk of death in those subjects above the 90th percentile in any measurement. In the logistic regression model, both low CAMA and low TSF were associated with a significantly increased risk of subsequent mortality. Poor nutritional state shown by low muscle bulk and fat stores was an important predictor of mortality, but obesity had no adverse effect on survival.