Weight-for-height indices of adiposity: Relationships with height in childhood and early adult life

Background: The expense and inconvenience of accurate assessment of fat/lean body mass have engendered a reliance on weight-for-height indices in epidemiological investigations; indices which are independent of height have been considered desirable. Methods: The relationship between weight-for-height and height was examined using the 1958 birth cohort, National Child Development Study, at ages 7, 11, 16, 23 and 33 years. For each age the sample was divided into a number of height groups; underweight, overweight and obesity were defined by relative weight (RW) and body mass index (BMI) in childhood and adulthood respectively. Results: In childhood the variance of RW showed substantial and systematic associations with height. Both underweight and overweight/obesity were related to height: patterns differed by age and sex, being most evident at age 7 in both sexes, continuing at age 11 (but more so in boys), and disappearing by age 16, At age 23, underweight was more prevalent and overweight and obesity less prevalent in the taller groups due to a linear correlation between BMI and height. At age 33 obesity was less prevalent in taller groups, particularly in women. Conclusions: These findings have implications for studies of obesity comparing groups which differ in height, for example, different cohorts or social classes. In the short term, interpretation of such results should take account of the phenomenon described. In the longer term, Information is needed on the relationship between height and more precise assessment of adiposity to confirm the findings of the current analysis.