This study examined differences in the incidence of wasting, stunting, and mortality among children aged 6 to 36 mo that received preventive supplementation with either RUSF or RUTF. To our knowledge, this is the first study to provide information on the relative performance of preventive supplementation strategies in young children using RUSF vs. RUTF to reduce the occurrence of malnutrition and mortality in young children. This study draws from an extensive surveillance database that included a relatively large number of children and high rates of follow-up (< 4% of follow-up visits missed). Application of propensity score methods to control for confounding by a number of measured factors allows the use of these unique data to inform the ongoing discussion on the use of RUSF within nutritional programs while randomized trial data become available.
This study, however, has several limitations. In addition to the dose, the two preventive strategies under comparison differ in important ways, including the duration of supplementation, mode and time of initiation of distributions, and age of children eligible for supplementation. The frequency of anthropometric screening also differed by strategy, with children receiving the RUSF strategy screened twice as often as those receiving the RUTF strategy owing to screening at both the RUSF distribution sites and monthly follow-up visits. As a result, our conclusions relate to the relative performance of the two preventive strategies overall, rather than the individual products. In addition, this comparison involves children from different districts. The study districts may differ with respect to baseline nutritional status, malaria endemicity, frequency of additional food aid distributions (e.g. corn-soy blend and oil), proximity to medical and nutritional care, and other unmeasured factors that influence the health and survival of children. While we are unable to ensure the comparability of children between districts owing to the non-randomized nature of the study, we do have information on a number of potential confounders. Adjustment for baseline anthropometry and other measured factors did not substantially alter our conclusions. Finally, we do not have complete data on compliance or supplement use within the household and thus cannot know if the supplement was consumed as intended by the target child.
We found that the effectiveness of preventive supplementation varied with the village experience with a previous nutritional intervention. The mechanisms underlying this interaction are unclear, but they are more likely related to contextual factors related to the village experience with the previous intervention than to individual factors associated with intake, such as baseline nutritional status. Children in villages where the previous nutritional intervention was implemented were of better nutritional status (measured by higher WHZ scores), and it is plausible that duration of supplementation may contribute more to improvements in weight gain than dose among children of better nutritional status. However, the effect of the supplementation strategy on the incidence of wasting or severe wasting was not modified by baseline WHZ or HAZ in supplemental analyses (data not shown). The interaction by previous intervention also persisted in the subgroup of children who were not eligible for the previous intervention due to their young age, again indicating that village-, rather than individual-level, factors associated with the previous intervention may contribute to the observed interaction.
In villages with previous experience with RUTF supplementation, RUTF may have been used as a replacement (as opposed to a complement) to habitual family meals or breast milk or shared with other household members. Either scenario could have contributed to lower energy intake with RUTF in villages where the previous intervention was implemented. Increased energy intake has previously been associated with increased weight gain (13
), and the energy provided by RUSF is within the range (200–300 kcal/day, assuming average breast milk intake) that older infants require from complementary foods (15
). Previous evaluations of RUSF supplementation have been consistent in demonstrating improved weight gain in a variety of study populations and against a range of comparator products, including micronutrient fortified flours and porridge (16
We found that the 6-mo RUSF strategy was related to a reduction in the incidence of stunting relative to the 4-mo RUTF strategy. It is possible that the duration, rather than the dose, of supplementation may contribute more to the maintenance of linear growth associated with the RUSF strategy. Although the impact of previous complementary feeding interventions on linear growth has been inconsistent (19
), ourstudy is consistent with the limited evidence specific to RUSF. RUSF was related to greater length gain compared to micronutrient-fortified flour among children aged 6–18 mo in Malawi (17
), a micronutrient-only supplement among older infants in Ghana (13
), and an unfortified spread among stunted children aged 3-6 y in Algeria (23
Owing to the interruption of the earlier trial, this study compared the performance of two preventive strategies in the context of whether or not a nutritional intervention was implemented in the previous year. The finding that previous intervention can modify the effectiveness of a nutritional program underscores that contextual factors should be considered early in program development, as the most effective dose and duration of supplementation may depend on the particular context of the program setting. Our findings suggest that there may be some settings in which there is no appreciable difference in the prevention of wasting between strategies that provide lower energy for longer duration and those that provide higher energy for shorter periods. However, the non-significant trend towards an increased risk of both wasting and severe wasting among children receiving the RUSF strategy in villages without the previous intervention is of concern. This should be confirmed in other studies.
Randomized trials that allow for direct estimation of the preventive effect of RUSF on the anthropometric and micronutrient status of young children are warranted. As age and nutritional status continue to be important predictors of nutritional outcomes, studies designed to compare the effectiveness of RUSF by age and nutritional status are also needed to identify groups in which supplementation is most effective and could be targeted. Finally, cost-effectiveness studies are required to help guide the choice of the strategy according to the context. While preventive strategies using RUSF for longer durations may be appropriate in some settings owing to its lower costs ($ 0.19/dose/day for RUSF vs. $ 0.37/dose/day for RUTF, written communication Stéphane Doyon, MSF, Paris, France, February 2009), the extended duration of such strategies will have additional indirect costs and programmatic implications.