Preventive small-quantity lipid-based nutrient supplements reduce severe wasting and severe stunting among young children: an individual participant data meta-analysis of randomized controlled trials
Kathryn G Dewey,Charles D Arnold,K Ryan Wessells,Elizabeth L Prado,Souheila Abbeddou,Seth Adu-Afarwuah,Hasmot Ali,Benjamin F Arnold,Per Ashorn,Ulla Ashorn,Sania Ashraf,Elodie Becquey,Kenneth H Brown,Parul Christian,John M Colford Jr,Sherlie Jl Dulience,Lia Ch Fernald,Emanuela Galasso,Lotta Hallamaa,Sonja Y Hess,Jean H Humphrey,Lieven Huybregts,Lora L Iannotti,Kaniz Jannat,Anna Lartey,Agnes Le Port,Jef L Leroy,Stephen P Luby,Kenneth Maleta,Susana L Matias,Mduduzi Nn Mbuya,Malay K Mridha,Minyanga Nkhoma,Clair Null,Rina R Paul,Harriet Okronipa,Jean-Bosco Ouédraogo,Amy J Pickering,Andrew J Prendergast,Marie Ruel,Saijuddin Shaikh,Ann M Weber,Patricia Wolff,Amanda Zongrone,Christine P Stewart
DOI: https://doi.org/10.1093/ajcn/nqac232
Abstract:Background: Meta-analyses show that small-quantity lipid-based nutrient supplements (SQ-LNSs) reduce child wasting and stunting. There is little information regarding effects on severe wasting or stunting. Objectives: We aimed to identify the effect of SQ-LNSs on prevalence of severe wasting (weight-for-length z score < -3) and severe stunting (length-for-age z score < -3). Methods: We conducted a 2-stage meta-analysis of individual participant data from 14 randomized controlled trials of SQ-LNSs provided to children 6-24 mo of age. We generated study-specific and subgroup estimates of SQ-LNS compared with control and pooled the estimates using fixed-effects models. We used random-effects meta-regression to examine study-level effect modifiers. In sensitivity analyses, we examined whether results differed depending on study arm inclusion criteria and types of comparisons. Results: SQ-LNS provision led to a relative reduction of 31% in severe wasting [prevalence ratio (PR): 0.69; 95% CI: 0.55, 0.86; n = 34,373] and 17% in severe stunting (PR: 0.83; 95% CI: 0.78, 0.90; n = 36,795) at endline. Results were similar in most of the sensitivity analyses but somewhat attenuated when comparisons using passive control arms were excluded (PR: 0.74; 95% CI: 0.57, 0.96; n = 26,327 for severe wasting and PR: 0.88; 95% CI: 0.81, 0.95; n = 28,742 for severe stunting). Study-level characteristics generally did not significantly modify the effects of SQ-LNSs, but results suggested greater effects of SQ-LNSs in sites with greater burdens of wasting or stunting, or with poorer water quality or sanitation. Conclusions: Including SQ-LNSs in preventive interventions to promote healthy child growth and development is likely to reduce rates of severe wasting and stunting. This meta-analysis was registered at www.crd.york.ac.uk/PROSPERO as CRD42019146592.