Are the measures used to calculate BMI accurate and valid for the use in older people?

Abstract

Body mass index (BMI) is widely accepted as the ‘gold standard’ for determining whether a patient is underweight or overweight. However, its use and accuracy is dependent on obtaining accurate measures of height and weight. Physiological changes, equipment variability and observer error leads to height being a measure fraught with inaccuracy. This has led to the development of other measures to calculate height, including demi‐span and knee height. But how much more accurate are these alternatives? The aim of this pilot study was to compare three anthropometric methods for measuring height in older people standing height, demi‐span and knee height.Seventeen subjects from a residential home in Leeds were chosen for the study, based on their ability to comply with the study protocol. Each subject was measured using a portable stadiometer for standing height, which was then adjusted for knee height, and a tape measure for demi‐span. All measurements were taken by the same researcher (TH). Measurements for demi‐span and knee height were converted using standard reference equations (for demi‐span: Bassey, 1986; for knee height: Chumlea, 1985). Data were analysed using SPSS. The Bland–Altman (1986) method for assessing agreement between two methods of measurements was carried out on the data.Data were obtained on 15 of the 17 subjects, as two subjects were unable to be weighed. Statistical analysis revealed that the lines of agreement between the three methods varied by up to 10 cm from the average or expected figure. In other words, there was up to 10 cm difference in height when calculated by the three different methods. In over half of the 17 subjects, an accurate measurement of standing height was difficult to obtain, because of difficulties encountered by the subjects in standing fully upright through frailty or spinal deformity. Problems were also encountered with demi‐span and knee height measurements, as some subjects had poor movement in their arms and were unable to stretch their arms out fully and 11 of 17 subjects were unable to get themselves into the correct position for measuring knee height.Although only conducted on a small sample of older people, the findings of this study highlight the difficulties in obtaining accurate height measurements for calculating BMI. While the literature suggests that demi‐span and knee height are reliable alternative methods for estimating stature, this was not the case in this study and may therefore also be a problem in clinical practice. Coupled with problems in the use of BMI in older people, the utility of performing such measures is debatable.