Risk of linear growth retardation during the first two years of life: a new approach.

OBJECTIVE: Estimate the risk of linear growth retardation during the first two years of life as a result of household social vulnerability. SETTING: Families who participated in the National Supplementary Feeding Program in the Health Units of the metropolitan area of the city of S?o Paulo, Brazil. SUBJECTS: Four hundred and thirty-one index-babies, weighing more than 2500 grams and who had at least one young sibling under the age of five who participated in the Program for a minimum of two years. DESIGN: The index-babies were divided into two cohorts: 74.9% coming from 'non-stunted families' (those with normal height siblings) and 25.1% from 'stunted families' (those with stunted siblings). The study design allowed the observation of growth patterns over a period of time and over a childhood growth range. It also allowed the estimation of the stunting and the recovery probabilities at each moment, not only within a given age range. The transition probabilities between 'stunted' and 'non-stunted' index-babies were estimated. The relative risk ratio (RR) was also calculated. RESULTS: The prevalence of stunting in the index-babies at 12 and 24 months of age was significantly greater in 'stunted families' (P < 0.001). Probabilities of becoming stunted began to differ from the fourth month on (confidence intervals non-superposed), and were higher for index-babies from 'stunted families'. The recovery probability of a stunted child was smaller in the 'stunted families' cohort after the 12th month of age. From the third month on, the (RR) was always above 1.5. CONCLUSION: The family context exposes children to failure in growth in the first two years of life when there are already stunted children in the household.