This is not the most recent version of the article. View current version (21 JUN 2013)

Intervention Protocol

You have free access to this content

Specially formulated foods for treating children with acute moderate malnutrition in low- and middle-income countries

  1. Marzia Lazzerini1,*,
  2. Laura Rubert1,
  3. Luca Ronfani2,
  4. Paola Pani1,
  5. Marcella Montico2

Editorial Group: Cochrane Developmental, Psychosocial and Learning Problems Group

Published Online: 18 JAN 2012

DOI: 10.1002/14651858.CD009584

How to Cite

Lazzerini M, Rubert L, Ronfani L, Pani P, Montico M. Specially formulated foods for treating children with acute moderate malnutrition in low- and middle-income countries (Protocol). Cochrane Database of Systematic Reviews 2012, Issue 1. Art. No.: CD009584. DOI: 10.1002/14651858.CD009584.

Author Information

  1. 1

    WHO Collaborating Centre for Maternal and Child Health, Institute for Maternal and Child Health, Unit for Health Services Research and International Health, Trieste, Italy

  2. 2

    Institute for Maternal and Child Health, Unit of Epidemiology and Biostatistics, Trieste, Italy

*Marzia Lazzerini, Unit for Health Services Research and International Health, WHO Collaborating Centre for Maternal and Child Health, Institute for Maternal and Child Health, IRCCS Burlo Garofolo, Trieste, 8232, Italy. lazzerini@burlo.trieste.it.

Publication History

  1. Publication Status: New
  2. Published Online: 18 JAN 2012

SEARCH

This is not the most recent version of the article. View current version (21 JUN 2013)

 

Background

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support
 

Description of the condition

Undernutrition encompasses chronic malnutrition (stunting), acute malnutrition (wasting), and deficiencies of micronutrients. Moderate acute malnutrition (MAM), also called moderate wasting, affects around 10% of children under five years of age in low- and middle-income countries (Black 2008). In absolute numbers, this means that between 40 to 55 million children in the world are suffering from MAM (Black 2008; Kerac 2011).

There is wide geographical variation in the prevalence of MAM in children under five years of age among low- and middle-income countries, and even within the same country. However, on average, South-Central Asia is estimated to have the highest point prevalence and the highest absolute number of affected children (that is a point prevalence of 19%, 30 million children affected). A prevalence rate above 15% is registered in several countries in Eastern, Middle, and Western Africa, and rates over 10% are reported in some countries in the Middle East (UNICEF 2011; WHO 2011).  

Time trends in the prevalence of MAM have shown important improvements in some regions of the world, notably Latin America (Fernandez 2002; Lima 2010), while in many other countries the prevalence still remains unacceptably high (UNICEF 2011; WHO 2011).  

The incidence of acute malnutrition in low- and middle-income countries is usually more prominent in the first years of life, when children have a high demand for nutrients and there are limitations in the quality and quantity of their diets, including inadequate breastfeeding practices (Shrimpton 2001; Black 2008; Paul 2011). Children in their first few years of life are also more susceptible to recurrent infectious diseases, such as diarrhoea, which adversely affect metabolism, appetite, and nutritional status. Being exposed to inadequate food intake and recurrent infections, children in low- and middle-income countries can easily enter a vicious circle of weight loss, increased susceptibility to infections, and ever–worsening nutritional status (Guerrant 2008). Acute malnutrition is also strongly associated with HIV and tuberculosis (TB). Both diseases compromise the nutritional status leading to malnutrition, which in turn aggravates the severity and the clinical course of HIV and TB (Donald 2007; Papathakis 2008).

Acute and chronic malnutrition can coexist in the same child. However, acute and chronic malnutrition are not necessarily associated on a geographical or ecological basis, that is, countries with a similar stunting prevalence can have a several-fold difference in the prevalence of wasting (Black 2008; UNICEF 2011; WHO 2011).

 
Consequences of moderate acute malnutrition

Moderate acute malnutrition (MAM) has a strong impact on child mortality and morbidity in developing countries. Malnutrition impairs immune function (Chandra 2002) and for this reason children with malnutrition are more vulnerable to infections, more prone to severe diseases, and present a higher mortality risk. Children with MAM have an estimated three- to four-fold increased risk of overall mortality compared to well-nourished children (Caulfield 2004; Black 2008). Cause-specific mortality risk in low- and middle-income countries for children with MAM is increased for common infections such as pneumonia (odds ratio (OR) 4.2; 95% confidence interval (CI) 3.2 to 5.5), measles (OR 3.7; 95% CI 2.5 to 5.5), malaria (OR 3.0; 95% CI 1.0 to 8.9), and diarrhoea (OR 2.9; 95% CI 1.8 to 4.5) (Black 2008). Moreover, children with MAM, if not adequately supported, can rapidly progress towards severe acute malnutrition (SAM), which is a life-threatening condition (Garenne 2009).  

In absolute numbers, most malnutrition-related deaths occur in children who are mildly or moderately malnourished, who are the larger proportion of total children, rather than in those who are severely malnourished (Pelletier 1995; Black 2008). It is estimated that about one in six (14.6%) of the child deaths per year in low- and middle-income countries are attributable to acute malnutrition. Of these, 4.4% are due to severe acute malnutrition, and 10.2% to moderate acute malnutrition. In terms of disease adjusted life years (DALYs), moderate and severe acute malnutrition together account for 14.8% of the total DALYs in children under five years of age. Investing in malnutrition is therefore essential to reduce child mortality and progress toward the Millennium Development Goal (MDG) 4 (Black 2008; Waage 2010). As acute malnutrition also increases the health risk associated with HIV and TB, interventions that aim to reduce malnutrition may also have an impact on these specific diseases (MDG 6) (World Bank 2006).

Malnutrition can adversely affect cognitive and social aspects of child health. Hunger is associated with reduced attention and low interest, and malnourished children have poor cognitive performances that can ultimately compromise their ability to learn, their education, and their overall development (World Bank 2006). Reaching an acceptable nutritional status is recognised as a fundamental prerequisite in order to improve educational achievement (MDG 2) (World Bank 2006; Stein 2008; Martorell 2010; Waage 2010).

From an economic perspective, malnutrition leads to direct losses in physical productivity, indirect losses from poor cognitive development and schooling, and a loss in resources from increased healthcare costs (World Bank 2006). Interventions that aim to reduce malnutrition have the potential to reduce poverty and to develop national economies (MDG1) (World Bank 2006; Waage 2010).

Malnutrition has a bidirectional relationship with social exclusion and poverty. The prevalence of malnutrition is often two or three times higher, sometimes even many times more, among those who are more socioeconomically deprived, even within the same geographical area (World Bank 2006; UNICEF 2011). The treatment of malnutrition is, therefore, also a matter of social justice and equity.

 

Description of the intervention

A framework developed by United Nation Children's Fund (UNICEF) recognises the basic and underlying causes of undernutrition, including environmental, economic, and sociopolitical contextual factors, with poverty having a central role (Black 2008). Addressing general deprivation and inequity would result in substantial and long-term reductions in undernutrition and should be a global priority.

Food interventions aim to reverse inadequate food intake, which is one of the immediate causes of malnutrition. Two broad approaches are used. In most situations, nutritional counselling on how to improve domestic diet is given to families on the assumption that they have access to all foods needed for feeding their children but lack the knowledge of how best to use them. In emergency contexts, where food availability is inadequate to meet the nutritional needs of children, improving local diet may not be practicable and externally provided food rations are given (Briend 2009).

Interventions to improve family foods include the use of special recipes made with locally available ingredients and home processing of foods, such as soaking, germination, malting, fermentation, aiming to increase their nutritional content. However, a recent review of programs implemented by a considerable number of United Nations agencies or donors, international non-governmental agencies (NGOs), paediatric associations, and local governments has highlighted that, in general, for the treatment of MAM there is a greater emphasis on providing food supplements than on improving locally available family foods (Ashworth 2009).

Externally provided food supplements include three main categories of foods, lipid-based nutrient supplements (foods with a high lipid content, characterised by high energy density and usually ready-to-use); blended food supplements (such as corn-soy blended, wheat-soy blended, sugar, oil, legumes or others); complementary food supplements (that is food-based complements to the diet that can be mixed with or consumed in addition to the diet).

Desirable characteristics of foods for children with MAM include adequate nutritional content, acceptability (that is taste and texture acceptable to children, cultural acceptability), low cost, easy preparation and administration in the context of resource-poor countries (Michaelsen 2009). In the emergency context, food would also need to be easily stored and distributed, and for this reason dry foods or special foods with low water content are usually preferred.

We do not have complete knowledge of which nutritional characteristics of foods for children with MAM are desirable in low- and middle-income countries. Nutritional requirements for children with moderate malnutrition have been recently reviewed (Golden 2009; Michaelsen 2009) and the World Health Organization (WHO) is in the process of defining recommendations on the nutritional composition of foods to rehabilitate children with MAM.

 

How the intervention might work

Food supplements given to children with acute malnutrition have the potential to rapidly improve the nutritional state of the children. On a population basis, improved access to adequate foods has the potential to reduce the prevalence of moderate and severe malnutrition, and the incidence of severe malnutrition.

A possible side effect of food supplements in children with MAM, in particular foods with high lipid content, is rapid weight gain, which has been recognised as a risk factor for adult adiposity and obesity (Ekelud 2006; Victora 2007). Homemade foods, especially if with high water content, may be at risk of being contaminated by germs and provoke other side effects such as diarrhoea.

On a population basis, externally provided supplementary feeding can be detrimental by creating dependence or, if ineffective, by representing a considerable waste of money.

 

Why it is important to do this review

There is no definitive consensus on the most effective way to treat children with MAM. To our knowledge, this is the first systematic review comparing all different types of foods for the treatment of children with MAM.

Previous narrative reviews have evaluated dietary counselling and other food interventions for children with MAM (Ashworth 2009; De Pee 2009). One review (Sguassero 2005) evaluated supplementary feeding but not directly in comparison with other foods as we will do in our review. One ongoing review is evaluating nutritional education as an intervention provided in addition to supplementary food and, for this reason, we will not address this specific question in our review (Sguassero 2007). Many other reviews have evaluated micronutrient interventions. For this reason the assessment of micronutrient intervention without food supplements is outside the scope of this review. One other review evaluated school feeding for school-aged children, which is an older age group than the one included in our review (Kristjansson 2007).

We expect to partially overlap with an ongoing review that is considering ready-to-use foods for all types of undernourished children, both SAM and MAM, and both stunting and wasting (Schoonees 2011). We considered it important to include this group of foods among the others examined in our review to allow us to assess all possible comparisons among the different types of existing foods that are used to rehabilitate children with MAM.

Many of the countries with a high burden of MAM are in a state of emergency or chronic hunger. Since the feasibility and acceptability of randomised controlled trials (RCTs) in emergency contexts is low, we will also include in the review studies with other designs, such as non-randomised controlled clinical trials (CCTs), controlled before-and-after studies (CBAs), and interrupted time series (ITS) studies.

 

Objectives

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

  1. To evaluate the safety and effectiveness of different types of foods for children with moderate acute malnutrition (MAM) in low- and middle-income countries
  2. To assess whether foods complying or not complying with specific nutritional compositions are safe and effective

 

Methods

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support
 

Criteria for considering studies for this review

 

Types of studies

Randomised controlled trials (RCTs), including cluster-randomised controlled trials (cRCTs), non-randomised controlled clinical trials (CCTs), controlled before-and-after studies (CBAs), and interrupted time series (ITS) studies.

There will be two minimum criteria for inclusion of CBAs.

  1. Contemporaneous data collection: we will include the study if the data in the experimental and control sites are collected in the same time frame.
  2. Appropriate choice of control site: we will include studies using a second site as control if the study and control sites are comparable with respect to the setting and population.

There will be two minimum criteria for inclusion of ITS designs.

  1. Clearly defined point in time when the intervention occurred: we will include the study if it is reported that the intervention occurred at a clearly defined point in time. If it is not reported in the paper and if the information cannot be obtained from the authors, we will exclude the study.
  2. At least three data points before and three after the intervention: we will include the study if three or more data points before and three or more data points after the intervention are recorded, and if a repeated measures analysis was done. If it is not specified in the paper, for example the number of discrete data points is not mentioned in the text or tables, and if information cannot be obtained from the authors, we will exclude the study. ITS studies that ignore secular (trend) changes and that performed a simple t-test of the pre- versus post-intervention periods without further justification will be excluded. If the ITS study did not provide a test for trend we will use the non-parametric test for trend across ordered groups, developed by Cuzick, which is an extension of the Wilcoxon rank-sum test.

 

Types of participants

Children in low- and middle-income countries aged six to 60 months with moderate acute malnutrition, treated either in hospital or in the community.

Moderate acute malnutrition (MAM) is defined as weight for height between -3 and -2 standard deviations from the mean or between 70% and 80% of the mean, or mid-upper arm circumference (MUAC) between 115 and 125 mm, and no oedemas. 

Children with very special needs, such as cerebral palsy or neoplasm, will not be included.

 

Types of interventions

 
Experimental

Any food used for children with moderate acute malnutrition. Any of the categories listed below will be included (De Pee 2009).

  1. Improved adequacy of local diet (such as local foods prepared at home according to a given recipe; home processing of local foods, such as soaking, germination, malting, fermentation).
  2. Lipid-based nutrient supplements (foods with high lipid content, characterised by a high energy density, and usually ready-to-use).
  3. Blended food supplements (such as corn-soy blended, wheat-soy blended, sugar, oil, legumes or others).
  4. Complementary food supplements (i.e. food-based complements to the diet that can be mixed with or consumed in addition to the diet). 

We anticipate that within each of these four intervention categories there will be large heterogeneity in the nutritional composition of the foods (De Pee 2009). Therefore foods will be further analysed by their nutritional composition. We will use the World Health Organization (WHO) nutritional recommendation for children with MAM as the standard. We will compare any food complying with the WHO recommendations with any food not complying. In this way, we will seek to assess whether foods complying or not complying with WHO recommendations are safe and effective, and whether the WHO recommendations are adequate.

Interventions providing micronutrients alone without extra calories will be excluded. Food interventions in the context of complex cross-sectional interventions such as food banks, conditional cash transfers, microcredit, and interventions promoting food production (such as home gardens and livestock farming) are outside the scope of this review.

 
Control

  1. Treatment as usual
  2. Alternative food

Concomitant interventions will be eligible only if administered concurrently to both the experimental and control groups.

 

Types of outcome measures

 

Primary outcomes

  1. Recovered
  2. Not recovered
  3. Progression to severe acute malnutrition
  4. Died
  5. Defaulted
  6. Weight gain
  7. Weight and height
  8. Mid-upper arm circumference (MUAC)
  9. Any adverse effect, including predisposition to obesity and diarrhoea

 

Secondary outcomes

  1. Nutritional adequacy of the diet
  2. Lean body mass increase
  3. Height gain
  4. Height and age
  5. Coverage of the population

 

Search methods for identification of studies

We will attempt to identify all relevant studies regardless of language or publication status (published, unpublished, in press, and in progress).

 

Electronic searches

 
Databases of published studies

We will search the following databases:

  • Cochrane Central Register of Clinical Trials (CENTRAL) (The Cochrane Library);
  • MEDLINE;
  • EMBASE;
  • LILACS;
  • CINAHL;
  • BIBLIOMAP;
  • POPLINE;
  • ZETOC.

The following search strategy will be used to search MEDLINE

1     Malnutrition/

2     Wasting Syndrome/

3     (malnutrition$ or malnourish$ or mal-nutrition$ or mal-nourish$).tw.

4     (wasting or stunting or growth-falter$).tw.

5     (undernutrition or undernourish$ or under-nutrition$ or under-nourish$).tw.

6     child nutrition disorders/ or infant nutrition disorders/ or childhood malnutrition/

7     Protein-Energy Malnutrition/

8     or/1-7

9     exp Dietary Supplements/

10     foods, specialized/

11     functional food/

12     Food, Fortified/

13     Food, Formulated/

14     Nutrition Therapy/

15     (food$ adj3 (complement$ or formulat$ or therap$ or supplement$ or fortif$ or blended or weaning)).tw.

16     ((nutrient$ or nutrition$) adj3 (complement$ or therap$ or supplement$)).tw.

17     (lipid based or (lipid adj3 supplement$) or LNS).tw.

18     ((home adj3 supplement$) or (home adj3 fortif$) or (home adj3 process$)).tw.

19     ("ready to use" or RUTF or RTUF or RUF or “plumpynut”).tw.

20     "point of use".tw.

21     Micronutrients/

22     (multimicronutrient$ or multi-micronutrient$ or micronutrient$ or micro-nutrient$ or multinutrient$ or multi-nutrient$).tw.

23     (MNP or MNPs or sprinkle$).tw.

24     or/9-23

25     (baby or babies or infant$ or child$ or toddler$ or preschool$ or pre-school$ or schoolchild$).tw.

26     exp Infant/

27     exp child/

28     or/25-27

29     8 and 24 and 28

We will modify search terms, as necessary, when searching other databases. We will not use any filters, such as limiting by language or to randomised controlled trials, in order to ensure that we do not miss any relevant study.

 
Databases of ongoing studies

  • WHO International Clinical Trials Registry Platform (ICTRP)
  • MetaRegister of Controlled Trials (mRCT)
  • ClinicalTrials.gov
  • United Nations System Standing Committee on Nutrition: Moderate Malnutrition e-platform

 
Other electronic searches

  • Field Exchange: the Emergency Nutrition Network Magazine

 

Searching other resources

 
Researchers, organizations

For unpublished and ongoing studies, we will contact a list of nutritional experts and researchers working in the field. This list will include contact persons working in the organizations and international groups reported below.

  • The World Health Organization (WHO); the United Nation Children's Fund (UNICEF); the World Food Program (WFP); the World Bank (WB); the United Nations Standing Committee on Nutrition (UNSCN); The United Nations Refugee Agency (UNHCR).
  • Technical bodies: the Food and Nutrition Technical Assistance Project (FANTA-2); the Emergency Nutrition Network (ENN); the International Malnutrition Task Force (IMTF); the Humanitarian Practice Network (HPN); the Community-Based Management of Acute Malnutrition (CMAM) Forum; the Global Nutrition Cluster (GNC); the Global Alliance for Improved Nutrition (GAIN).
  • Academic institutions: the International Centre for Diarrhoeal Disease Research (ICDDR); the Institute of Child Health London (ICH); the Medical School Blantyre/Mangochi Malawi; the University California Davis; the Washington University at St. Louis; the London School of Hygiene and Tropical Medicine (LSHTM); and the Institute of Tropical Medicine (ITP) Antwerp, Belgium.
  • International non-government organizations (NGOs): Save the Children (SC); Doctors without Borders (MSF); Valid international; Concern Worldwide; Action Against Hunger (ACF); and others.

 
Conference proceedings

  • Commonwealth Association for Paediatric Gastroenterology And Nutrition (CAPGAN) meeting, 21 to 23 July 2011, London, UK.  

 
Reference lists

We will also check the reference lists of all studies identified by the above methods.

 

Data collection and analysis

 

Selection of studies

Two review authors (ML and LRU) will independently screen the titles and abstracts yielded by the search against the inclusion criteria listed above. We will obtain the full text of papers or reports for studies that appear relevant, or for which more information is needed to determine their relevance. The same authors will independently screen the papers to determine whether they meet the criteria for inclusion. We will resolve any disagreement about eligibility through discussion and, when disagreements cannot be resolved, by seeking advice from the other authors (MM, LRO). We will seek additional information from the authors of the studies, as necessary, to resolve questions about the relevance or methodology of a trial. We have the capacity within our team to deal with studies in English, Portuguese, Spanish, and French. If we have studies in other languages, we will liaise with the editorial base to have them translated, where necessary. We will document the reasons for excluding studies. None of the authors will be blind to the authors, institutions, or the journals of publication of the articles.

 

Data extraction and management

Two authors (ML and LRU) will independently extract data for each trial using a data extraction form that has been piloted in order to collect information about the population, the setting, the intervention, the outcome measures, the process, and the risks of bias. If data are missing or unclear, we will attempt to contact the trial authors. To avoid mistakes due to data manipulation, we will first collect the data as they are reported and, if any transformation is needed, we will transform them subsequently. We prefer to use endpoint scores rather than changes from baseline (Higgins 2008). When both unadjusted and adjusted effect estimates are available, we will create a standardised mean difference effect size by employing adjusted means and unadjusted standard deviations. To calculate unadjusted standard deviations from given adjusted standard deviations we will use the covariate-outcome correlation. If more than one adjusted mean is given, we will choose the estimate that is identified as the primary adjusted model by the authors or, if this is not specified, the model that adjusted for the maximum number of covariates.

Two authors (PP and ML) will assess each included study in order to determine the nutritional content of the intervention and its conformity with WHO recommendations (WHO 2011). If the nutritional content is not specified in the original article, it will be calculated using a nutrient analysis software system (MICRODIET).

For survival outcomes, if data are available we will transform time-to-event data into dichotomous data. If the status of patients is not known at a fixed time point, we will analyse the survival data separately.

 

Assessment of risk of bias in included studies

Two review authors (ML and LR) will independently rate the quality of each study. We will use the Cochrane 'Risk of bias' tool for RCTs (Higgins 2008), modified with the Cochrane Effective Practice and Organisation of Care Group (EPOC) criteria (EPOC 2009). The risk of biases in the included studies will be summarised in 'Risk of bias' tables.

 
Risk of bias criteria for RCTs, CCTs, CBAs

 
Random sequence generation

  • Low risk: if the investigators describe a random component in the sequence generation process (such as referring to a random number table; using a computer random number generator; coin tossing, shuffling cards or envelopes, throwing dice, drawing of lots; minimisation).
  • High risk: if the investigators describe a non-random component in the sequence generation process. Usually the description would involve some systematic, non-random approach (for example sequence generated by odd or even date of birth; sequence generated by some rule based on date of admission; sequence generated by some rule based on hospital or clinic record number; allocation by judgement of the clinician; allocation by preference of the participant; allocation based on the results of a laboratory test or a series of tests; allocation by availability of the intervention). CCTs, CBAs, and ITS will be scored as 'high risk'.
  • Unclear risk: insufficient information about the sequence generation process to permit judgement.

 
Allocation concealment 

  • Low risk: participants and investigators enrolling participants could not foresee assignment because one of the following, or an equivalent method, was used to conceal allocation, central allocation by telephone, web-based and pharmacy-controlled randomisation; sequentially numbered drug containers of identical appearance; sequentially numbered, opaque, sealed envelopes.
  • High risk: participants or investigators enrolling participants could possibly foresee assignments and thus introduce selection bias (for example allocation based on using an open random allocation schedule such as a list of random numbers; assignment envelopes were used without appropriate safeguards, such as if envelopes were unsealed or non­opaque, or not sequentially numbered; alternation or rotation; date of birth; case record number; any other explicitly unconcealed procedure). CBAs and ITSs will be scored as 'high risk'.
  • Unclear risk: insufficient information to permit judgement. This is usually the case if the method of concealment is not described or not described in sufficient detail to allow a definite judgement (for example if the use of assignment envelopes is described but it remains unclear whether the envelopes were sequentially numbered, opaque, and sealed).

 
Blinding of participants and personnel

  • Low risk: blinding of participants and key study personnel ensured and unlikely that the blinding could have been broken; no blinding or incomplete blinding but the review authors judge that the outcome is not likely to be influenced by lack of blinding.
  • High risk: no blinding or incomplete blinding and the outcome is likely to be influenced by lack of blinding; blinding of key study participants and personnel attempted but likely that the blinding could have been broken, and the outcome is likely to be influenced by lack of blinding.
  • Unclear risk: insufficient information to permit judgement; the study did not address this outcome.

 
Blinding of outcome assessment

  • Low risk: no blinding of outcome assessment but the review authors judge that the outcome measurement is not likely to be influenced by lack of blinding; blinding of outcome assessment ensured and unlikely that the blinding could have been broken.
  • High risk: no blinding of outcome assessment and the outcome measurement is likely to be influenced by lack of blinding; blinding of outcome assessment but likely that the blinding could have been broken, and the outcome measurement is likely to be influenced by lack of blinding.
  • Unclear: insufficient information to permit judgement; the study did not address this outcome.

 
Incomplete outcome data 

We will extract and report data on attrition and exclusions, as well the numbers involved (compared with total number randomised), reasons for attrition or exclusion (where reported or obtained from investigators), and any re-inclusions in analyses performed by the review authors.

  • Low risk: no missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size. Missing data have been imputed using appropriate methods.
  • High risk: reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size; 'as-treated' analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation.
  • Unclear: insufficient reporting of attrition and exclusions to permit judgement (for example number randomised not stated, no reasons for missing data provided); the study did not address this outcome.

 
Selective reporting 

  • Low risk: the study protocol is available and all of the study’s prespecified (primary and secondary) outcomes that are of interest in the review have been reported in the prespecified way; the study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were prespecified (convincing text of this nature may be uncommon).
  • High risk: not all of the study’s prespecified primary outcomes have been reported; one or more primary outcome is reported using measurements, analysis methods or subsets of the data (for example subscales) that were not prespecified; one or more reported primary outcome was not prespecified (unless clear justification for the reporting is provided, such as an unexpected adverse effect); one or more outcome of interest in the review is reported incompletely so that it cannot be entered in a meta-analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study.
  • Unclear: insufficient information to permit judgement. It is likely that the majority of studies will fall into this category.

 
Other bias 

  • Low risk: the study appears to be free of other sources of bias.
  • High risk: there is at least one important risk of bias. For example, the study had a potential source of bias related to the specific study design used, or has been claimed to have been fraudulent, or had some other problem.
  • Unclear: there may be a risk of bias but there is either insufficient information to assess whether an important risk of bias exists or insufficient rationale or evidence exists that an identified problem will introduce bias.

 
Additive risk of bias criteria for RCTs, CCT, CBAs

 
Baseline characteristics

  • Low risk: if baseline characteristics of the study and control providers are reported and are similar.
  • High risk: if there is no report of characteristics in the text or tables, or if there are differences between control and intervention providers.
  • Unclear: if it is not clear in the paper (for example characteristics are mentioned in the text but no data were presented).
  • If some primary outcomes were imbalanced at baseline, assessed blindly, or affected by missing data, and others were not, each primary outcome can be scored separately.

 
Protection against contamination

  • Low risk: if allocation was by community, institution, or practice and it is unlikely that the control group received the intervention.
  • High risk: if it is likely that the control group received the intervention (for example if patients rather than professionals were randomised).
  • Unclear: if professionals were allocated within a clinic or practice and it is possible that communication between intervention and control professionals could have occurred (for example physicians within practices were allocated to intervention or control).

 
Additive risk of bias criteria for ITS

 
Was the intervention independent of other changes?

  • Low risk: if there are compelling arguments that the intervention occurred independently of other changes over time and the outcome was not influenced by other confounding variables or historic events during the study period. We will note if any events or variables are identified.
  • High risk: if it is reported that the intervention was not independent of other changes over time.

 
Was the shape of the intervention effect prespecified?

  • Low risk: if the point of analysis is the point of intervention or a rational explanation for the shape of the intervention effect was given by the author(s). Where appropriate, this should include an explanation if the point of analysis is not the point of intervention.
  • High risk: if it is clear that the condition above is not met.

 
Was the intervention unlikely to affect data collection?

  • Low risk: if reported that the intervention itself was unlikely to affect data collection (for example sources and methods of data collection were the same before and after the intervention).
  • High risk: if the intervention itself was likely to affect data collection (for example any change in source or method of data collection reported).

We will exclude ITS studies that ignore secular (trend) changes and that performed a simple t-test of the pre- versus post-intervention periods without further justification.

 

Measures of treatment effect

 

Dichotomous data

Where dichotomous data are presented, we will record the number of participants experiencing the event in each group, and a risk ratio with a 95% confidence interval will be calculated for each outcome in each trial (Higgins 2008).

 

Continuous data

We will analyse continuous data when means and standard deviations are presented in the study papers, made available by the authors of the studies, or are calculable from the available data. Where outcomes are measured using the same scale, we will calculate a mean difference to determine the differences in mean scores between groups. Where similar outcomes are measured using different scales, we will calculate a standardised mean difference using Hedges g.

 

Time-to-event data

We will present the treatment effects of time-to-event data, or survival data (for example child maltreatment incidence data), as a hazard ratio with 95% confidence interval.

 

Unit of analysis issues

 

Cluster-randomised trials

For cluster-randomized trials (cRCT), we will follow the methods for adjusting for clustering as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008). If a cRCT has properly accounted for the cluster design, it can be included in a meta-analysis by simply using the effect estimate and its standard error (SE) and using the generic inverse variance method in Review Manager 5 (RevMan 2011). If an appropriate analysis has not been performed, the cRCT will be incorporated into a meta-analysis (if relevant) by using an 'approximate method'. This entails calculation of an 'effective sample size' for the comparison groups by dividing the original sample size by the 'design effect', which is 1 + (c-1)ICC, where c is the average cluster size and ICC is the intra-cluster correlation coefficient. For dichotomous data, both the number of participants and the number experiencing the event will be divided by the same design effect, while for continuous data only the sample size needs to be reduced (means and standard deviations (SDs) should be left unchanged). If available, we will extract the required information from the articles; otherwise we shall attempt to contact the study authors. If we fail to obtain the required information, we will perform sensitivity analyses using different ICC values. Although the values are relatively arbitrary, we prefer to use these to adjust the effect estimates and their standard errors (SEs) due to the implausibility that the ICC is actually 0. We will then combine the estimates and their corrected SEs from the cRCT with those from parallel group designs using the generic inverse variance method in Review Manager 5.

 

Multiple interventions per individual

If the participants in some studies receive a food plus other treatments, we will also meta-analyse the studies separately and, if useful, this characteristic will be considered in the meta-regression. The discussion of these results will take into account the additional treatments received (Higgins 2008).

 

Studies with multiple treatment groups

For studies where there are multiple treatment groups, we will not analyse data from the same group twice. We will select the treatment condition for meta-analysis according to which one matches the inclusion criteria. The comparison condition will be treatment-as-usual or alternative foods.

 

Multiple time points

We will report data at all collection time points during and after the intervention period (follow-up). Where data allow, we plan to group the time points as follows: less than three months of food treatment, three to six months of food treatment, and more than six months of food treatment.

 

Dealing with missing data

We will assess missing data and dropouts in the included studies. We will investigate and report the reasons, numbers, and characteristics of dropouts. We will make efforts to contact the authors when further information or data are necessary. In any meta-analyses, we will use data from all original participants when possible, and we will report when that is not the case. For studies in which the missing data are not available, we will conduct sensitivity analyses to assess potential bias in the analysis and discuss the extent to which the results might be biased by missing data.

 

Assessment of heterogeneity

We will examine heterogeneity among included studies through the use of the Chi2 test, where a low P value indicates heterogeneity of treatment effects. We will also use the I2 statistic (Higgins 2002) to determine the percentage of variability that is due to heterogeneity rather than to sampling error or chance. We will discuss the possible reasons for any heterogeneity. We may use subgroup analyses and meta-regression to further investigate heterogeneity, as described below.

 

Assessment of reporting biases

 

Assessment of reporting biases

We will use funnel plots to investigate the relationship between effect size and standard error, when possible. When such a relationship is found, we will examine clinical diversity as a possible explanation. Asymmetry could be attributed to publication bias or related biases.

 

Data synthesis

The methods we use to synthesise the studies will depend on their design and heterogeneity. We will group studies and describe them according to setting, type of intervention, and study design. We will consider only groups of studies of the same design for pooling (that is RCTs, CBAs, and ITS will be grouped separately).

We will use a meta-analysis if at least two studies are homogeneous regarding the participants, interventions, and outcomes. If it is not possible to combine outcome data due to differences in the reporting or substantive heterogeneity, we will report results in tables and the text. We will describe ITS studies in the text and summarise them in a table, used only for descriptive purposes. As a certain degree of clinical heterogeneity is expected, we will use a random-effects meta-analysis. If an adjusted effects estimate is used, generic inverse variance will be chosen as the method of meta-analysis.

If we have enough RCTs, indirect comparison methods and a combination of direct and indirect comparisons in a multiple-treatment meta-analysis (MTM) will be used in the case that the assumptions needed to apply these methods (Higgins 2008; Salanti 2008) are properly satisfied. 

 

Subgroup analysis and investigation of heterogeneity

Conducting a large number of subgroup analyses increases the likelihood of false positive results, and therefore it is important to carefully select in advance the relevant characteristics to be investigated (Higgins 2008). Eight characteristics are identified and we feel that all of them are of significant clinical importance:

  1. level of food security;
  2. breastfeeding practices (children breastfed versus non-breastfed);
  3. age (≤ 2 years versus > 2 years);
  4. stunting (children with stunting versus without stunting);
  5. HIV or TB (children with HIV or TB versus those without HIV or TB);
  6. acute complications (children with acute complications versus children without acute complications);
  7. prevalence of wasting in the population (< 10% versus 10% to 15% versus > 15%);
  8. prevalence of stunting in the population (< 30% versus 30% to 39% versus > 40%);
  9. prevalence of HIV or TB in the population (high versus medium versus low).

These analyses will be exploratory as they involve non-experimental (cross-over) comparisons, and we will treat any conclusion with caution.

If useful and appropriate, we will also conduct meta-regression to look at the relationship of size of effect to the above reported characteristics of the studies, plus process characteristics such as: case definition of MAM (for example using MUAC versus weight-for-height); duration of the intervention (if useful, in combination with intensity of the intervention); micronutrient contents; presence of co-interventions (for example education); acceptability of food given (taste, variety, cultural acceptability); compliance with the intervention; and coverage (for studies at a population level).

 

Sensitivity analysis

If the methodology or analyses in the studies might conceivably have affected the robustness of the results of the review, we wil conduct sensitivity analyses by removing studies with particular characteristics and re-analysing the remaining studies to determine whether the relevant factors affect the results. These analyses will be conducted to examine the effects of:

  1. the removal of studies with high risk of bias;
  2. changing the way that values are imputed for missing data;
  3. for cRCTs only with missing ICC, reanalysing the data using different ICC values;
  4. reanalysing the data using different statistical approaches (for example using a fixed-effect model instead of a random-effects model) (Higgins 2008).

             

 

Acknowledgements

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

We thank the members of the WHO Nutrition Guidance Expert Advisory Group (NUGAG) for their suggestions in the design of this review.

 

History

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

Protocol first published: Issue 1, 2012

 

Contributions of authors

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

All authors contributed to the development of this protocol. ML drafted the protocol; LRO and MM provided statistical assistance, and LRU provided extra input. ML and LR devised the search strategy. ML and LRU will screen the abstracts and titles and retrieve potentially eligible papers. ML and LRU will review the papers and make decisions about eligibility in discussion with LR and MM. ML and LRU will extract data. PP will check and calculate the nutritional contents of foods. ML will draft the full review with regular input from all authors at every stage.

 

Declarations of interest

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support

  • Marzia Lazzerini (ML) - none known
  • Laura Rubert (LRU) - none known
  • Luca Ronfani (LRO) - none known
  • Paola Pani (PP) - none known
  • Marcella Montico (MM) - none known

 

Sources of support

  1. Top of page
  2. Background
  3. Objectives
  4. Methods
  5. Acknowledgements
  6. History
  7. Contributions of authors
  8. Declarations of interest
  9. Sources of support
 

Internal sources

  • Institute for Child Health, IRCCS Burlo, Trieste, Italy.

 

External sources

  • Department of Nutrition for Health and Development, WHO, Geneva, Switzerland.

References

Additional references

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Acknowledgements
  7. History
  8. Contributions of authors
  9. Declarations of interest
  10. Sources of support
  11. Additional references
Ashworth 2009
  • Ashworth A, Ferguson E. Dietary counseling in the management of moderate malnourishment in children. Food and Nutrition Bulletin 2009;30 Suppl(3):405-33.
Black 2008
  • Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 2008;371(9608):243-60.
Briend 2009
  • Briend A, Prinzo ZW. Dietary management of moderate malnutrition: time for a change. Food and Nutrition Bulletin 2009;30 Suppl(3):265-6.
Caulfield 2004
  • Caulfield LE, de Onis M, Blössner M, Black RE. Undernutrition as an underlying cause of child deaths associated with diarrhea, pneumonia, malaria, and measles. American Journal of Clinical Nutrition 2004;80(1):193-8.
Chandra 2002
De Pee 2009
  • De Pee S, Bloem MW. Current and potential role of specially formulated foods and food supplements for preventing malnutrition among 6- to 23-month-old children and for treating moderate malnutrition among 6- to 59-month-old children. Food and Nutrition Bulletin 2009;30 Suppl(3):434-63.
Donald 2007
  • Donald P, Maher D, Qazi S, World Health Organization. A research agenda for childhood tuberculosis. Improving the management of childhood tuberculosis within national tuberculosis programmes: research priorities based on a literature review. Geneva: WHO, 2007.
Ekelud 2006
  • Ekelund U, Ong K, Linné Y, Neovius M, Brage S, Dunger DB, et al. Upward weight percentile crossing in infancy and early childhood independently predicts fat mass in young adults: the Stockholm Weight Development Study (SWEDES). American Journal of Clinical Nutrition 2006;83(2):324-30.
EPOC 2009
  • Cochrane Effective Practice and Organisation of Care Review Group (EPOC). Risk of bias. Data collection checklist. Available from: www.epoc.cochrane.org/epoc-author-resources 2009.
Fernandez 2002
  • Fernandez ID, Himes JH, de Onis M. Prevalence of nutritional wasting in populations: building explanatory models using secondary data. Bulletin of the World Health Organization 2001;80(4):282-91.
Garenne 2009
  • Garenne M, Willie D, Maire B, Fontaine O, Eeckels R, Briend A, et al. Incidence and duration of severe wasting in two African populations. Public Health Nutrition 2009;12(11):1974-82.
Golden 2009
  • Golden MH. Proposed recommended nutrient densities for moderately malnourished children. Food and Nutrition Bulletin 2009;30 Suppl(3):267-34.
Guerrant 2008
Higgins 2002
Higgins 2008
  • Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. Chichester: John Wiley & Sons, 2008.
Kerac 2011
  • Kerac M, Blencowe H, Grijalva-Eternod C, McGrath M, Shoham J, Cole TJ, et al. Prevalence of wasting among under 6-month-old infants in developing countries and implications of new case definitions using WHO growth standards: a secondary data analysis. Archives of Disease in Childhood Feb 2011 [Epub ahead of print]; Vol. 96, issue 11:1008-13.
Kristjansson 2007
  • Kristjansson B, Petticrew M, MacDonald B, Krasevec J, Janzen L, Greenhalgh T, et al. School feeding for improving the physical and psychosocial health of disadvantaged students. Cochrane Database of Systematic Reviews 2007, Issue 1. [DOI: 10.1002/14651858.CD004676.pub2]
Lima 2010
  • Lima AL, Silva AC, Konno SC, Conde WL, Benicio MH, Monteiro CA. Causes of the accelerated decline in child undernutrition in northeast Brazil (1986-1996-2006). Revista de Saude Publica 2010;44(1):17-27.
Martorell 2010
  • Martorell R, Horta BL, Adair LS, Stein AD, Richter L, Fall CHD, et al. Weight gain in the first two years of life is an important predictor of schooling outcomes in pooled analyses from five birth cohorts from low- and middle-income countries. Journal of Nutrition 2010;140(2):348–54.
Michaelsen 2009
  • Michaelsen KF, Hoppe C, Roos N, Kaestel P, Stougaard M, Lauritzen L, et al. Choice of foods and ingredients for moderately malnourished children 6 months to 5 years of age. Food and Nutrition Bulletin 2009;30 Suppl(3):343-404.
MICRODIET
  • Downlee Systems Ltd. MICRODIET. Salford, UK: Downlee Systems Ltd, 1999.
Papathakis 2008
  • Papathakis P, Piwoz E. Nutrition and Tuberculosis: A review of the literature and considerations for TB control programs. Washington, DC: USAID, 2008.
Paul 2011
  • Paul VK, Sachdev HS, Mavalankar D, Ramachandran P, Sankar MJ, Bhandari N, et al. Reproductive health,and child health and nutrition in India: meeting the challenge. Lancet 2011;377(9762):332-49.
Pelletier 1995
  • Pelletier DL, Frongillo EA Jr, Schroeder DG, Habicht JP. The effects of malnutrition on child mortality in developing countries. Bulletin of the World Health Organization 1995;73(4):443-8.
RevMan 2011
  • The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager 5. 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011.
Salanti 2008
Schoonees 2011
  • Schoonees A, Lombard M, Nel E, Volmink J. Ready-to-use therapeutic food for treating undernutrition in children from 6 months to 5 years of age. Cochrane Database of Systematic Reviews 2011, Issue 2. [DOI: 10.1002/14651858.CD009000]
Sguassero 2005
  • Sguassero Y, de Onis M, Carroli G. Community-based supplementary feeding for promoting the growth of young children in developing countries. Cochrane Database of Systematic Reviews 2005, Issue 4. [DOI: 10.1002/14651858.CD005039.pub2]
Sguassero 2007
  • Sguassero Y, Booker D, Dennis JA, Orellano A, Abalos E. Supplementary feeding with nutritional education for caregivers for promoting growth and development in young children in developing countries. Cochrane Database of Systematic Reviews 2007, Issue 2. [DOI: 10.1002/14651858.CD006509]
Shrimpton 2001
Stein 2008
  • Stein AD, Wang M, DiGirolamo A, Grajeda R, Ramakrishnan U, Ramirez-Zea M, et al. Nutritional supplementation in early childhood, schooling, and intellectual functioning in adulthood: a prospective study in Guatemala. Archives of Pediatrics and Adolescent Medicine 2008;162(7):612-8.
UNICEF 2011
  • Unicef Childinfo. Monitoring the Situation of Women and Children. www.childinfo.org/index.html (accessed March 2011).
Victora 2007
Waage 2010
  • Waage J, Banerji R, Campbell O, Chirwa E, Collender G, Dieltiens V, et al. The Millennium Development Goals: a cross-sectoral analysis and principles for goal setting after 2015. Lancet and London International Development Centre Commission. The Lancet 2010;376(9745):991-1023.
WHO 2011
  • World Health Organization. WHO Global Databases on Child Growth and Malnutrition. www.who.int/nutgrowthdb/en/ (accessed March 2011).
World Bank 2006
  • The World Bank. Repositioning Nutrition as Central to Development. A Strategy for Large-Scale Action. Washington, DC: The International Bank for Reconstruction and Development/The World Bank, 2006.