Abdominal circumference contributes to absence of wasting in Brazilian children

Juraci A. César, Cesar G. Victora, Saul S. Morris, Cora A. Post

Research output: Contribution to journalArticlepeer-review

9 Scopus citations


A number of population groups in Latin America show high prevalences of stunting (low height-for-age) despite very low rates of wasting (weight-for- height deficits). One possible explanation for this phenomenon is an increase in abdominal circumference, which would affect children's weights but not their heights. This study was designed to describe the abdominal circumferences of a group of poor children from Northeast Brazil, and to relate these to their weight-for-weight z-score. Children (n = 252) participating in a government growth monitoring program were studied. The prevalence of stunting (below -2 SD) was 26.2%, but only 1.2% were wasted. Abdominal circumferences increased with age up to 36 mo, followed by a slight decline after 48 mo. Abdominal circumference was the anthropometric measurement most closely associated with weight-for-height, with a coefficient of determination of 41%. Even after adjusting for arm circumference, abdominal circumference continued to explain 16% of the variation in weight-for-height. Despite slight differences in measurement techniques, the study children had consistently larger abdominal girths than a sample of North American children. These findings must be verified by replication but highlight a possible contribution of abdominal circumference in the determination of levels of wasting.

Original languageEnglish (US)
Pages (from-to)2752-2756
Number of pages5
JournalJournal of Nutrition
Issue number11
StatePublished - Nov 1996
Externally publishedYes


  • abdominal circumference
  • Brazil
  • malnutrition
  • preschool children
  • wasting

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Food Science


Dive into the research topics of 'Abdominal circumference contributes to absence of wasting in Brazilian children'. Together they form a unique fingerprint.

Cite this