No significant difference in weight gain after correction for cluster design
To the editor
In a trial featured on the cover of the BMJ Alderman et al report on
gain in 27,995 children who were cluster randomized in 48 parishes to
albendazole for treatment of soil-transmitted helminths or nothing (1).
We have recently published a Cochrane review of the effects on growth
school performance of deworming for soil-transmitted helminths in children
and included this study (2). The paper reported a weight gain of 2.413 kg
the 25 treatment parishes and 2.259 kg in the control parishes at an
unspecified follow-up point. The 154 g difference was reported as
statistically significant in the paper. We contacted the authors as the
did not make it clear if the design effects had been taken into account in
result. The authors wrote that the result was unadjusted, and kindly
us with adjusted data: these results showed no significant difference
for the primary outcome of weight gain between intervention and control
groups (difference 154 g; 95% CI -19.7 to 330 g).
We think it is important that the correction outlined above - which
published (2) - is linked to the published version of this trial.
1. Alderman H, Konde-Lule J, Sebuliba I, Bundy D, Hall A. Effect on
gain of routinely giving albendazole to preschool children during child
days in Uganda: cluster randomised controlled trial. BMJ
2. Taylor-Robinson DC, Jones AP, Garner P. Deworming drugs for
soil-transmitted intestinal worms in children: effects on growth and
performance. Cochrane Database of Systematic Reviews 2007, Issue 4. Art.
No.: CD000371. DOI: 10.1002/14651858.CD000371.pub3.
Competing interests: No competing interests