Stunting during infancy and schooling: a prospective study in Senegal

The objective was to investigate the relationship of stunting in late infancy with school achievement in a rural area of Senegal. A cohort of 4753 subjects born 1987‐95 for whom length‐for‐age at the age of 9 mo had been measured during vaccination was followed for school attendance from 2001 to 2009. Factors associated with outcomes (any school attendance, delayed entry, delay in grades, dropout) were analyzed using multiple logistic regression. Potential confounders were year of birth, sex, level of schooling of parents, economic level and size of village. The prevalence of stunting at age 9 mo was 17.5%. One third of subjects never attended school. The risk was independent of early stunting (P>0.20) but closely associated with year of birth, schooling of parents, economic level and size of village (P<0.001 for all). Delayed school entry (蠅8 y) was more common among stunted children (adjusted OR: 1.7, 95% CI: 1.1, 2.4, P<0.01) and in smaller villages (2.4, CI: 1.8, 3.2) and less common if the father had secondary education (0.2, CI: 0.0, 0.8). Delay in grades 蠅3y was 28% and 53%, respectively, in grades 4 and 6. Stunted children were at greater risk (1.5, CI: 1.2, 1.9, P=0.001 and 1.6, CI: 1.2, 2.0, P<0.001, respectively), whereas girls and those whose parents had secondary education had lower risks. Girls also had lower risks of dropout before entering the secondary level (0.70, CI: 0.6, 0.9, P<0.01) whereas stunting was not associated with dropout. Progression through grades was independent of stunting (1.00, 0.75, 1.33). In conclusion, children who were stunted during infancy entered school later than other children and did not catch up in grades during primary school. Delay in grades was a universal phenomenon in this population.