The WHO international growth reference was reviewed to identify characteristics that might affect the interpretation of anthropometric indicators derived from it. Discontinuities were found at the junction point of the two distinct data sets from which the reference curves are derived. Also, the Z-score derived percentile curves for weight-for-age and weight-for-height did not join due to differences in the variances of the two portions of the reference. Empirically an examination of data from international nutrition surveys revealed an artifactual drop in the prevalence of low height-for-age, low weight-for-height, and high weight-for-height at age 2 y, which is due substantially to these discontinuities rather than to intrinsic changes in nutritional status with age. Strategies are discussed for the interpretation of anthropometric indicators in light of these characteristics of the international reference.