RT Journal
A1 Roberfroid D, Huybregts L, Kolsteren P
T1 EFfect of preventive supplementation on young children in niger
JF JAMA
JO JAMA
YR 2009
FD June 3
VO 301
IS 21
SP 2208
OP 2209
DO 10.1001/jama.2009.739
UL http://dx.doi.org/10.1001/jama.2009.739
AB 
First, the authors used different reference distributions to define malnutrition at inclusion in the study (National Center for Health Statistics/World Health Organization [NCHS/WHO], 1978) and at inclusion in the analysis (WHO Child Growth Standards, 2006). A proportion of participants were thus excluded from the analysis on the grounds that they were already malnourished at recruitment according to the WHO reference.2 This proportion was likely substantial and would be expected to consist mainly of children younger than 24 months because the 2 references yield different diagnoses principally in that age range.2 Unfortunately, this age range has the highest risk of malnutrition.3 This compromises the internal validity of the study, and extrapolation of results to children aged 6 to 60 months requires caution. The problem could be amplified by the interaction of the intervention with child age at baseline (P = .07), a result not fully discussed by the authors.