Measuring catch-up growth in malnourished populations.

BACKGROUND AND AIM Numerous recent studies measure catch-up growth by regressing adult or pre-adolescent height on early childhood height. Using simple statistical theory and data from a healthy and well-nourished population, this paper shows that these tests are uninformative about the extent of catch-up growth. The study also provides new empirical evidence on pubertal catch-up growth using longitudinal data for rural Tanzania. SUBJECTS AND METHODS The 1970 British Cohort Study is used to demonstrate the flaws in the recent literature using regression techniques to study catch-up growth. The data for the empirical analysis come from the Kagera Health and Development survey-a longitudinal study spanning two decades. The final sample includes 540 children whose heights are measured in early childhood and in adulthood. Catch-up growth is measured as the change in height-for-age z-score over time. RESULTS The mean HAZ-score in the cohort improves from -1.86 in early childhood to -1.20 in adulthood. Without catch-up growth, children would have been 4.5-5 centimetres shorter adults. Graphical analysis shows that most of this catch-up growth takes place in puberty. CONCLUSION Catch-up growth after early childhood is possible. Puberty seems to offer an opportunity window for recovery.