Biostatistics Research Today is a free monthly online journal that collates and summarizes the latest research about Biostatistics, including details on statistics, uncertainty, probability, modeling.

Ghrelin, adiponectin, and leptin do not predict long-term changes in weight and body mass index in older adults: longitudinal analysis of the Rancho Bernardo cohort.

Langenberg C, Bergstrom J, Laughlin GA, Barrett-Connor E

Department of Family and Preventive Medicine, School of Medicine, University of California, San Diego, La Jolla, CA 92093-0607, USA.

Ghrelin, leptin, and adiponectin are associated with body size in cross-sectional studies; it is unknown whether these hormones predict long-term changes in body size. Multilevel models were used to study associations between fasting serum hormones, measured in 698 men and 619 women (60-91 years) in samples collected at baseline (1984-1987), and changes in weight and body mass index, assessed repeatedly over a follow-up period of up to 18 years (median, 4.7 years). Baseline weight was -1.5 kg lower for a one-standard-deviation increment in ghrelin and -3.3 kg lower for a one-standard-deviation increment in adiponectin, similar in men and women. For leptin, baseline weight was 12.1 kg higher for a one-standard-deviation increment in men, compared with 5.7 kg in women (sex-interaction p < or = 0.0001). Ghrelin and adiponectin did not affect weight change; their associations with weight were constant over time, indicated by nonsignificant hormone-by-time interactions. The positive association between leptin and weight became slightly weaker over time. Results were similar when investigating repeated measures of body mass index. From this analysis of Rancho Bernardo Study data, the authors conclude that ghrelin, adiponectin, and leptin do not predict weight gain beyond reflecting the influence of attained body size on future changes in weight or body mass index.

Published 7 December 2005 in Am J Epidemiol, 162(12): 1189-97.
Full-text of this article is available online (may require subscription).

© 2005-2008 Biostatistics Research Today. All Rights Reserved.