ze published studies that analyze the effects of long-term change in body weight on all-cause mortality. The thirteen reports are from eleven diverse population studies, seven from the United States (including the Baltimore Longitudinal Study of Aging) and four from Europe. All studies included a weight change period of four years or more followed by a mortality assessment period of eight years or more. Changes in weight occurred during the adult years of life, the youngest starting age being 17 years. Study details varied widely in terms of numbers of subjects and deaths, ages at initial and final weight measurements, duration of the mortality follow-up period, consideration of cigarette smoking and other potential confounders, exclusion criteria, temporal separation between the weight change and mortality follow-up periods, and the method of assessment of the association of weight change and all-cause mortality. Conclusions. Despite the diversity of (a) populations studied, (b) degree of """"""""clinical clean-up"""""""" at entry, (c) techniques for assessing weight change, and (d) differences in analytical techniques, including consideration of potentially confounding variables, certain conclusions may be drawn. The preponderance of the evidence is that during the adult years, highest mortality occurs in those who either have lost weight or gained excessive weight and lowest mortality is generally associated with modest gain in weight. These results support the use of age-specific weight-for-height tables, as recommended initially by the Gerontology Research Center, and as supported subsequently by the 1989 National Research Council publication the 1990 Dietary Guidelines for Americans (USDA ind USDHHS).
