The issue of the effect of age on the recommended range of weights-for-height remains controversial. The major controversy involves the question of adjustment of recommended weights for age. The widely used Metropolitan Tables, most recently revised in 1983, are said to be applicable to ages 25 to 59 years. In the past year, a re-analysis of data from the insurance industry as published in Build Study 1979 has been initiated and comprehensive collation of all pertinent studies from the world's literature has been accomplished. The reanalysis was necessary consequent to the publication of a study on the adult population of Norway in 1984 by Waaler. That study showed clearly that the relationship between the Body Mass Index (BMI, wt/ht2) and the mortality rate, although U-shaped, was not symmetrically so. The upswing in mortality in increasingly underweight subjects was sharper than the upswing in the increasingly overwieght groups. The Norway Study is of major importance (1) because of its size and (2) because it includes men and women from 20-24 to 85-89 years of age. We have shown that the asymmetric data distribution could be well fit by a curve obtained by computing the log-log BMI vs. the mortality ratio. Examination of the Build Study 1979 data showed again that there was a tendency to similar asymmetry in the U-shaped relationship. For this reason these data were re-computed as noted for the Norway Study. It was found that the fit was indeed better. This may require re-computation of recommended weights, but the impact will be relatively small: the upper weight limits of our previously computed age-specific table may need to be adjusted upward slightly and the lower weight limits may need to be adjusted upward slightly in younger subjects and downward in older subjects. The Norway data are very similar to the Build Study data. Over the comparable age range of 20-69 years, there is less than 1% overall difference despite the large differences in the types of populations studied.
