Discriminant analysis by anthropometric measures in elderly Bengalee Hindus of Calcutta, India

Although a sizeable proportion of India's population is elderly (55 years and above), virtually no information exists on how well anthropometric measures could be utilized to correct age group size in elderly individuals. The present investigation was aimed to study how good anthropometric measures could be manoeuvre for corrected group size in the elderly Bengalee Hindu population of Calcutta, India. A cross-sectional study was undertaken among 410 elderly (55 years and above) Bengalee Hindu individuals (men = 210, women = 200) utilizing various anthropometric measures, namely height, weight, body mass index (BMI), mid upper arm circumference (MUAC) and triceps skinfold. A discriminant analysis was then undertaken separately for men and women using the above mentioned anthropometric measures. The results revealed that overall 77.1% for men and 78.6% for women of all cases were correctly classified. No individuals were misclassified in the predicted groups III and I for both the sexes. The most discriminating variable for the groups was triceps skinfold in both sexes. These results provided evidence that the anthropometric characteristics of group III and I were very different from each other with those belonging to group II having intermediate characteristics.