Community-based versus health facility-based management of acute malnutrition for reducing the prevalence of severe acute malnutrition in children 6 to 59 months of age in low- and middle-income countries

  • Protocol
  • Intervention



This is the protocol for a review and there is no abstract. The objectives are as follows:

  1. To compare the outcomes of uncomplicated SAM in children under five years old treated through the 'out-patient therapeutic programme (community-based approach)' with those treated with the 'in-patient care at a therapeutic feeding centre' approach in low- and middle-income countries'.

  2. To compare the effectiveness of both the community-based approach and in-patient care at therapeutic feeding centres in treating uncomplicated SAM in low- and middle-income countries in both humanitarian and non-humanitarian situations


Description of the condition

Malnutrition is responsible for more than half of the 10 million deaths annually among children under the age of five in low- and middle-income countries. Both chronic and acute malnutrition are now considered as major threats to public health in these countries (Black 2003; Caulfield 2002; Rice 2000). Sub-Saharan Africa and South-Asia have the highest prevalence, with approximately 73 million children under five suffering from acute malnutrition (Collin 2006a; Schofield 1996). Acute malnutrition includes undernutrition (low weight-for-age) and wasting (low weight-for-height), and is associated with serious morbidity and high mortality (Cauldwell 2004; Collin 2006; Collin 2006a; Pelletier 2003). Severe acute malnutrition (SAM) is a serious and major challenge to achieving the Millenium Development Goals for reducing mortality in under-fives (Mason 2003; Pelletier 2003; UNICEF 2010; WHO 2007). Globaly, around 20 million under-five children are suffering from severe acute malnutrition, mostly in Sub-Saharan Africa and Asia. (UN 2007) The prevalence of SAM is as high as 6% in Pakistan and 2.8% in India. Similarly, in these countries, 1-1.5 million deaths (mortality rate 73 to 187 per 1000 per year) are associated with this condition annually (Andre 2006; Collin 2006, UN 2007).

SAM or wasting is predominantly measured by one or more of the following criteria:

  1. Weight-for-height (WFH) less than -3 Z-scores (i.e. three standard deviations, or more, below the mean);

  2. Weight-for-height less than 70% of the median;

  3. Mid-upper-arm circumference (MUAC) less than 11.5 cm or 115 mm;

  4. Presence of bilateral pitting oedema (SPHERE 2004; WHO 1999, WHO 2006).

In this review we will include only children with uncomplicated SAM, aged from six months to 59 months, with any of the above criteria. SAM requires a very specialised and organised set up of management protocols, treatment guidelines and preventive measure to achieve control and reduction of prevalence in low- and middle-income countries (Collin 2006a).

Management of uncomplicated SAM was based on World Health Organization (WHO) standard protocol(s) that included individual in-patient care, nutritional rehabilitation, medical treatments and family counselling (Carlos 2002; SPHERE 2004; WHO 1999; WHO 2006). Conventionally, uncomplicated SAM cases are managed at the health facility level, commonly known as 'therapeutic feeding centres' (Grobler-Tanner 2004; Khanum 1994), although patients may require admission to the health facility for the entire duration of treatment. Normally, length of stay would be around three weeks, however, this could be longer if the child remains unstable (Bachmann 2010; Gatchell 2006). Prolonged hospital stays utilise more resources in terms of cost, and impose a burden that eventually limits the programme coverage, and, ultimately, reduces beneficial impact on the population (Grobler-Tanner 2004). However, this model has been considered to be very successful because it has been associated with good clinical outcomes and reduced SAM-associated mortality (ENN/FANTA 2008; Khara 2004; Collin 2003).

The health-facility management model, however, has received some criticism in the past few decades; one of its major weaknesses is that it does not take into account certain important factors in hospitalisation of these sick children. Notably, these include socioeconomic factors, such as: workload of women in a rural setting; family size, and other children at home (Carlos 2002); usual health care-seeking behaviour of parents and access of community to health-care services; all of which play a key role in deciding whether or not a child is treated for certain illnesses (Babar 2008). There is also a huge burden on tertiary care, especially in resource-poor settings where primary- and secondary-care level facilities are under utilised (Babar 2008; Carlos 2002).

Furthermore, this in-patient care requires parents or caregivers to stay with the child until complete rehabilitation is achieved (Grobler-Tanner 2004). This can be too much of a burden in a situation where a child has uncomplicated SAM. Such cases can be managed and rehabilitated at community level, without creating any burden on tertiary-care settings or on families. Referral to a tertiary-care setting may have a financial impact on poor families in terms of out-of-pocket expenditure during the hospital stay (Carlos 2002; Khara 2004). In addition, hospitalisation exposes the severely and acutely malnourished child to nosocomial infections (hospital-acquired infections), which can complicate the hospital treatment (Carlos 2002; Khara 2004).

Description of the intervention

In the last few decades, there has been a paradigm shift in the management of acute malnutrition from a facility-based to community-centred approach utilising mobile teams and primary-care health centres (Ashworth 2001; Collin 2001). The modern community approach is usually known as 'community-based management of acute malnutrition' (ENN/FANTA 2008). The focus of this latest approach is to treat the majority of acutely malnourished children in the community utilising the primary health services. The aim of this approach is to reduce the unnecessary referral burden on tertiary or in-patient care, which is frequently overcrowded by acutely malnourished children without complications - cases that can be treated and managed easily at the community level, without referral to hospital (Collin 2006; Deconink 2008).

The community-based approach uses a network of out-patient treatment sites to provide a take-home food ration known as 'ready-to-use therapeutic food' along with essential medicines (7 days course oral Amoxicillin) as prophylaxis against pneumonia and other respiratory illnesses in uncomplicated SAM case. (Carlos 2002; ENN/FANTA 2008; Manary 2004). The results of community-based management of acute malnutrition are satisfactory; overall it produces improved clinical outcomes and reduced mortality rates in acutely malnourished children (Collin 2003,Collin 2003). Community-based management of acute malnutrition involves direct community engagement. It provides a framework for an integrated public-health response to address acute malnutrition by managing, treating and preventing deterioration in most children with a moderate level of acute malnutrition and uncomplicated SAM at the household level, and reserving in-patient care for complicated cases (Deconink 2008) (Figure 1). Its decentralised approach minimises geographical barriers to access to care in both humanitarian and non-humanitarian situations (Collin 2001; Deconink 2008; UN 2008).

Figure 1.

Community-based management of acute malnutrition approach (Deconink 2008)

How the intervention might work

The WHO had classified acute malnutrition into moderate and severe malnutrition based on anthropometric measures and the presence of bilateral pitting oedema (Collin 2003; Deconink 2008; WHO 1999). With the paradigm changing from facility-based to community-based management of acute malnutrition, the classification, referral criteria and management protocol were re-visited. The new classification now splits acute malnutrition into 'moderately malnourished', 'SAM without complications' and 'SAM with complications'. Referral criteria have been re-designed according to this new classification in order to be operationally compatible with community-based management of malnourished children in middle- and low-income countries (Collin 2003). (The new classification is presented in detail in Figure 2) Hence, by using this new strategy proposed by the WHO, the United Nations Children's Fund (UNICEF), Save the Children etc, a large number of children with uncomplicated SAM can easily be managed at the community level, without being admitted to a health facility or a therapeutic feeding centre (Allen 2001). The community-based management approach encompasses timely detection of SAM in children, and ensures provision of treatment for those without medical complications (Collin 2002).

Figure 2.

'Classification and admission criteria of different types of acute malnourished children to different nutritional programmes of community-based management of acute malnutrition (Collin 2003)

Community-based management has four core components that address the basic referral system according to the classification of SAM:

  1. 'Community mobilisation or outreach' is a broad component that focuses on strengthening the role of traditional leaders, healers and other people within the community to enhance mobilisation, participation and engagement in the programme, thus maximising beneficial outcomes of the intervention (Bandawe 2003; Collin 2003).

  2. The 'supplementary feeding program', focuses on moderately malnourished children and uses standard protocols to manage them with ready-to-use supplementary food. This component has mostly mobile health coverage, and provides follow-up treatment at home, within the community, or, in some cases, at the health facility.

  3. The 'out-patient therapeutic program' rehabilitates and follows children with uncomplicated SAM in the community within existing primary health-care infrastructure at static sites (Collin 2003). The main focus is to manage the child with high energy food supplements, ready-to-use therapeutic food and other community-based protocols for treatment.

  4. 'Stabilisation centres' are specialised centres that provide in-patient care for stabilisation and rehabilitation of children with SAM and its complications, and identify and treat life-threatening problems, specific deficiencies and metabolic abnormalities, using standard feeding formulas of F100/F75 and treatment protocols (Bachmann 2010; Collin 2003; WHO 1999). Once a child meets the discharge criteria, its follow-up will be conducted at out-patient treatment sites and through other components of the community-based management system, until it reaches the point of complete recovery (Collin 2006).

Why it is important to do this review

There are a few fundamental questions that will form the core of this review. Firstly, is not clear at present how the outcomes compare for children with uncomplicated SAM treated through the out-patient component of the community-based approach compared to those treated through the old health facility-based approach.

Secondly, at present large investments are being used to promote community-based management of acute malnutrition in low- and middle-income countries, especially during humanitarian crises. However, whether the community-based approach is applicable and feasible within routine health programmes in non-emergency situations is not clear. There are strong recommendations to scale-up the community-based approach for children with uncomplicated SAM. Very few studies have looked at the effectiveness of community-based management of acute malnutrition compared to the traditional approach of facility-based treatment for uncomplicated SAM. Community-based care with ready to-use therapeutic food has been suggested as being a virtuous way to confront uncomplicated SAM in low- and middle-income settings, but the efficacy, effectiveness, and cost-effectiveness of this modern approach compared to the traditional approach of a health facility based in poor health districts with local staff is still unproven (Ashworth 2001; Ashworth 2006; Collin 2006).

Lastly, other aspects, including the coverage and effective implementation of the community-based approach, and a review of existing programmes run by routine health services, need to be compared with traditional approaches in order to generate the most robust evidence on uncomplicated SAM (Ciliberto 2005; Collin 2006), therefore, we want to look at both approaches from every critical aspect of managing uncomplicated SAM.

Comparison of management of uncomplicated SAM through a health facility-based approach against a community-based approach is imperative, and there is a strong need for policy direction for all the key stakeholders who may require more robust evidence for the decisions required to scale-up community-based management of acute malnutrition at national level, and to integrate it into the existing health system.


  1. To compare the outcomes of uncomplicated SAM in children under five years old treated through the 'out-patient therapeutic programme (community-based approach)' with those treated with the 'in-patient care at a therapeutic feeding centre' approach in low- and middle-income countries'.

  2. To compare the effectiveness of both the community-based approach and in-patient care at therapeutic feeding centres in treating uncomplicated SAM in low- and middle-income countries in both humanitarian and non-humanitarian situations


Criteria for considering studies for this review

Types of studies

We will include a variety of study designs in this review based on the criteria set down by the Cochrane Effective Practice and Organisation of Care (EPOC) Group:

  1. Randomised controlled trials (RCTs): any experimental design where children suffering from SAM without complications are allocated to one or other of the interventions i.e. community-based management of acute malnutrition or facility based management. Studies with only one intervention will be excluded.

  2. Cluster-randomised trials: studies with at least two intervention sites and two comparator sites will be included.

  3. Controlled before-and-after studies: the timing of the period of the study in both the intervention and comparator should be comparable. Pre- and post-intervention periods of measurement of both groups will be the same. Both groups will be comparable for key characteristics.

  4. Non-randomised controlled trials (NRCTs): studies with at least two intervention sites and two comparator sites will be included.

  5. Interrupted time-series studies: studies with a clearly defined point in time when the intervention occurred; these studies must have at least three data points before and after the intervention.

  6. We will also consider repeated measure trials if, within interrupted time-series studies, we find that repeated measures are made in the same population (SAM without complications) at each time point.

We will also include a specific unpublished report (audit reports, monitoring and evaluation reports and the like) and grey literature in this review. If reports do not meet the included studies criteria but are relevant, they will be included in the review as the part of an appendix to guide those who are interested in knowing what interventions could be assessed further using more rigorous designs and methodology. We will also try to provide a guidance framework for those who are interested in knowing how the intervention could be implemented successfully in a specific situation in low-and middle-income countries.

Types of participants

The participants of this review will be children aged six to 59 months with uncomplicated SAM. Uncomplicated SAM is predominantly measured by (WHO 1999; WHO 2006, SPHERE 2004):

  1. Weight-for-height less than -3 Z-scores (i.e. three standard deviations, or more, below the mean);

  2. Weight-for-height less than 70% of the median;

  3. Mid-upper-arm circumference (MUAC) less than 11.5 cm or 115 mm;

  4. Presence of bilateral pitting oedema;

  5. No other complication.

Types of interventions


Community-based management of uncomplicated SAM.


Health facility-based management of uncomplicated SAM.

The third component of community-based management of acute malnutrition, the 'out-patient therapeutic programme', focuses on managing uncomplicated severely malnourished children with high energy food supplements; 'ready-to-use therapeutic food' and other community based-protocols of treatment, such as prophylaxis with amoxicillin, without referring the child to the tertiary-care setting. This component is mainly run by physicians and healthcare workers trained in the management of malnourished children. This component is basically a treatment protocol with rigorous follow-up of the child at community level using mobile health units or the set-up of primary health care units within the community as a unit for this component. For this review this third component will remain focused.

Conversely, the health-facility approach will use in-patient care for uncomplicated severely malnourished children. Children must be admitted for management and follow the standard treatment guidelines. For this review we are specifically focusing on comparison of these two specific approaches of managing uncomplicated cases of SAM.

Types of outcome measures

Primary outcomes

We will consider the following two primary outcomes:

  1. Percentage reduction in the prevalence of children with uncomplicated SAM in the target community where the intervention was implemented;

  2. cure rate from uncomplicated SAM at 12 weeks in a specific intervention.

We define cure rate as the number of children who have completely recovered from uncomplicated SAM per 1000 children with uncomplicated SAM.

The complete recovery of children with uncomplicated SAM will be assessed according to standard anthropometric measurement criteria:

  1. Weight-for-height is more than 90% for two consecutive out-patient therapeutic programme visits;

  2. MUAC more than 115 mm for two consecutive out-patient therapeutic programme visits;

  3. No oedema for two consecutive out-patient therapeutic programme visits;

  4. Clinically well.

Secondary outcomes
  1. Percentage of defaults: (a) number of children with SAM without complications who were absent for two or three consecutive visits (out-patient therapeutic programme visits occur either every two weeks, or every week); or (b) in the case of those receiving health-facility management, number who leave against medical advice.

  2. Death: deaths from SAM without complication in a specific intervention at any point when treated at out-patient therapeutic programme or in a health facility.

  3. Not recovered: number of children who have not achieved the cure criteria within four months.

  4. Resource use under each intervention and comparator, such as out-of-pocket expenditures.

  5. Burden (e.g. work load) on healthcare providers.

  6. Burden on the families in terms of time.

  7. Utilisation of services, coverage rate or access to specific intervention.

  8. Equity aspect (fairness) of both the intervention and comparator where the programme is implemented.

  9. Complications: percentage of children with uncomplicated SAM admitted to out-patient therapeutic programme (community-based) or in-patient care (health facility-based) later, having developed any kind of complication(s), which may include:

    • anorexia, no appetite for ready-to-use therapeutic food;

    • vomiting;

    • hypothermia (body temperature lower than or equal to 35.5°C);

    • fever (body temperature higher than or equal to 38.5°C);

    • severe pneumonia;

    • severe dehydration;

    • severe anaemia;

    • not alert (very weak, lethargic, unconscious, fits or convulsions);

    • conditions requiring intravenous infusion or nasogastric tube feeding.

Percentage reduction will best be measured in controlled before-and-after studies and interrupted time-series designs. If studies provide intervention and control groups, and further before and after effects, then we will also measure the difference in reduction. We will also include studies based on specified outcomes for this review, as defined above. There are a few outcomes that are best measured using specific study designs; we will include the studies accordingly.

Search methods for identification of studies

Electronic searches

We will search the following electronic databases for primary studies

  • the Cochrane Central Register of Controlled Trials (CENTRAL), including the Cochrane EPOC Group Specialised Register;

  • MEDLINE In-Process & Other Non-Indexed Citations and MEDLINE (1946 to present) (Ovid);

  • EMBASE (1980 to present) (Ovid);

  • CINAHL (1980 to present) (EBSCOHost);

  • Global Health (CAB Direct);

  • Global Health Library (AIM (AFRO), IMEMR (EMRO), IMSEAR (SEARO) and WPRIM (WPRO)) (WHO);

  • LILACS (Virtual Health Library);

  • WHOLIS - the WHO Library Information System;

  • Health Management (1986 to present) (ProQuest);

  • Science Citation Index, Social Sciences Citation Index (1975 to present) (ISI Web of Science).

A search strategy for MEDLINE has been developed by the Cochrane EPOC Group Trials Search Co-ordinator in consultation with the authors. See Appendix 1 for the full search strategy. Strategies for other databases will be based on the MEDLINE strategy and translated appropriately for each database.

Related systematic reviews will be identified by searching the Database of Abstracts of Reviews of Effectiveness (DARE, The Cochrane Library).

Searching other resources

Grey literature

We will search for grey literature via the New York Academy of Medicine Library (

Trial registries

We will search the following registers for ongoing trials:

Additional studies

We will search for additional studies and reports from the official web library of organisations with extensive working experience of nutritional interventions. We will also try to meet the focal person at selected organisations.

We will contact researchers and organisations with a nutrition wing, such as the WHO, UNICEF, the United States Agency for International Development (USAID), Save the Children, Merlin, and Médecins Sans Frontières; Science Citation Index and Social Sciences Citation Index (ISI Web of Science) will also be searched for studies that cite any of the studies that will be included in the review.

Data collection and analysis

Selection of studies

Two authors will screen titles and abstracts of studies for inclusion, then will retrieve the full text of potentially-eligible studies for screening, and, independently, will apply the inclusion criteria to those retrieved publications. We will discuss any disagreements about the inclusion of studies, and, if no consensus is met, we will consult the contact editor of the review. We will seek further information from the authors where papers contain insufficient information to make a decision about eligibility.

Data extraction and management

We will extract data from the included studies using a pre-standardised data extraction form. Two authors will independently extract all outcome data. We will then cross-check the data, and, if necessary, will make reference to the original paper. Any outstanding discrepancies between the two data extraction sheets will be discussed by the data extractors and resolved by consensus. We will also contact study authors to obtain any missing information. We will extract data relating to the following from all the included studies:

  1. Population: specifically, data will be extracted about the children (male and female) with uncomplicated SAM aged between six months and 59 months. We will also consider baseline anthropometry, if it has been done in any study for this population group.

  2. Intervention: including data on the 'out-patient therapeutic component' of the community-based approach. We will consider adding data from other components, such as the inclusion of 'community mobilisation', 'supplementary feeding programme' and 'stabilisation centre' during analysis, if they are found to be significantly relevant to the review. Also, data will be extracted about the intervention-related funding source; the implementing partner (International and National non-governmental organisations); the degree and level of implementation; the degree and level of integration within existing government health; and duration of intervention. Also, with regard to the implementing partner of intervention (i.e. who delivered the Intervention): details about how the programme's quality procedures were followed (e.g. training of staff, supervision and monitoring).

  3. Comparator: data will be extracted on the children with uncomplicated SAM admitted to in-patient care for treatment and rehabilitation. We will not include studies that focused on in-patient care for children with complicated cases of SAM.

  4. Outcomes: we will extract data pertaining to the primary and secondary outcomes defined earlier. Studies that only address our secondary outcomes will also be included in this review.

  5. Setting: data will be collected on the setting, context and type of intervention; whether delivered in humanitarian crisis or famine; any co-intervention such as education or access to mass media, or growth monitoring exercises; the energy level of the food and type of therapeutic food with regard to macro- and micronutrient levels.

Assessment of risk of bias in included studies

Two authors will independently assess each included study's risk of bias using a form with the standard criteria described by the Cochrane EPOC Group (EPOC 2009). We will use the Cochrane EPOC Group's nine-point criteria for RCTs, NRCTs, and controlled before-and-after studies, and the seven-point criteria for interrupted time-series studies to determine the quality of all eligible studies. When information available from the studies is not sufficient, we will attempt to contact the study authors and request further details. We will not exclude studies on the grounds of their quality, but will clearly report methodological quality when presenting the results of the studies.

We will perform further analysis of the quality of evidence on primary outcomes using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach (Guyatt 2008). Using this approach, we will rate the quality of the body of evidence for each key outcome as 'high', 'moderate', 'low', or 'very low'.

Criteria for randomised and non-randomised controlled trials
  1. Was the allocation sequence adequately generated?

  2. Was the allocation adequately concealed?

  3. Were baseline outcome measurements similar? *

  4. Were baseline characteristics similar?

  5. Were incomplete outcome data adequately addressed? *

  6. Was knowledge of the allocated interventions adequately prevented during the study? *

  7. Was the study adequately protected against contamination?

  8. Was the study free from selective outcome reporting?

  9. Was the study free from other risks of bias?

Trials will be awarded a 'Yes' if there is no evidence of other risk of biases. If studies are awarded an 'Unclear' or a 'No', but there are sufficient data in the paper to do an adjusted analysis (e.g. baseline adjustment analysis or Intention-to-treat analysis) the criteria should be adjusted to 'Yes'.

* 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.

Criteria for interrupted time-series studies
  1. Was the intervention independent of other changes?

  2. Was the shape of the intervention effect pre-specified?

  3. Was the intervention unlikely to affect data collection?

  4. Was knowledge of the allocated interventions adequately prevented during the study?

  5. Were incomplete outcome data adequately addressed?

  6. Was the study free from selective outcome reporting?

  7. Was the study free from other risks of bias?

Measures of treatment effect

We will record and report measures of effect in the same way that the investigators have reported them. If possible, we will standardise measures of effect as mean differences (MD) in natural units or use a standardised scale to allow for comparisons across studies, where between-study comparisons are relevant. For interrupted time-series analyses, we will report the immediate effect after the specified transition period, at one year, and, if reported, after longer periods of follow-up.

For dichotomous data, we will present the results as risk ratios (RRs) with 95% confidence intervals (CIs). For continuous data, we will use the MD if outcomes were measured in the same way between trials. We will use the standardised mean difference (SMD) to combine trials that measured the same outcome, but used different methods.

Unit of analysis issues

We will include cluster-randomised trials in the analyses along with individually randomised trials. Their sample sizes will be adjusted by the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), using a design effect reported from the trial. Alternatively, it will be estimated using an estimate of the intra-cluster correlation co-efficient (ICC) derived from the trial (if possible), from a between-cluster coefficient of variation provided in the published paper, for a similar trial or from a study of a similar population by the formula: design effect = 1 + (average cluster size - 1) x ICC (Higgins 2011).

Dealing with missing data

We will note the level of attrition for included studies. We will attempt to obtain missing data from the investigators. If this is not possible we will report the data as missing, and will not attempt to impute values. In cases where authors have reported on controlled before-and-after and interrupted time-series studies, but have conducted inappropriate analyses on their data, we will re-analyse these data, provided the necessary values are published. For those that have been analysed or reported using inadequate statistical methods, we will analyse outcomes separately from both these studies.

Assessment of heterogeneity

We will determine the clinical heterogeneity, and, wherever possible, we will assess statistical heterogeneity in each meta-analysis using the T2, I2 and Chi2 statistics. We will regard heterogeneity as substantial if I2 is greater than 30% and either T2 is greater than zero, or there is a low P value (less than 0.10) in the Chi2 test for heterogeneity.

During the review process, we will identify several factors that might explain heterogeneity (please refer to section on subgroup analysis). These along with a priori factors will be undertaken as exploratory, hypothesis-generating analyses.

We anticipate that the eligible studies will exhibit significant heterogeneity due to variations in target outcomes that we will need to evaluate for both the intervention and comparator, such as types of population and provider(s) of the services. Methodological features, characteristics of the interventions and the contexts in which interventions are delivered are also important. Meta-regression is one approach that addresses these variations. Given the number of potentially relevant covariates, however, meta-regression would require many more studies than we anticipate finding. There is also the possibility that most of the eligible studies would assign an intervention status to the provider, rather than the population under study, but would not take cluster effects in the analysis into account (i.e. they would exhibit unit of analysis errors). Performing either a conventional meta-analysis or meta-regression using studies with unit of analysis errors would require us to make a number of assumptions about the magnitude of unreported parameters, such as the intra-class correlation coefficients and the distributions of children across clusters, in order to avoid spurious precision in 95% CIs.

To preserve the goal of providing a quantitative assessment of the effects associated with computerised reminders, without resorting to numerous assumptions or conveying a misleading degree of confidence in the results, we will report the median improvement in process adherence (and interquartile range) among studies that share specific features of interest. This approach was first developed in a large review of strategies to foster the implementation of clinical practice guidelines (Grimshaw 2004), and subsequently applied to reviews of quality improvement strategies in a series of reports for the US Agency for Healthcare Research and Quality (Shojania 2004a; Shojania 2004b; Walsh 2006). This method of reporting the median effect sizes across groups of studies involves two distinct uses of the term 'median' (Shojania 2009).

First, in order to handle multiple outcomes within individual studies, for each study we will calculate the median improvement in process adherence across the various outcomes reported by that study. We will also capture instances when a study identified a primary outcome and analyse those studies separately. Then we will perform a sensitivity analysis in which, instead of the median outcome, we used the best outcome from each study. With each study then represented by a single, median outcome, we will then calculate the median effect size and interquartile range across all included studies. It is this second use of the median that is crucial to the method. Instead of providing a conventional meta-analytic mean (an average weighted on the basis of the precision of the results from each study), we highlight the median effect achieved by included studies, along with an interquartile range for these effects.

Assessment of reporting biases

We will assess selective outcome reporting as a risk of bias criterion, as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will assess the risk of publication bias qualitatively based on the characteristics of the included studies. For example, if we identify only a few studies that indicate effects in favour of the interventions, this would raise our concern about the risk of publication bias.

Data synthesis

We anticipate that the breadth of the interventions included in this review will be great, making it difficult to combine all intervention results in statistical analysis. However, if the results of subsets of the studies can be pooled, we will do so in meta-analyses.

We will carry out statistical analysis using the Cochrane Collaboration's statistical software, Review Manager 2013. We will use a fixed-effect analysis for combining data where it is reasonable to assume that studies are estimating the same underlying treatment effect, i.e. where trials are examining the same intervention, and the trials' populations and methods are judged to be sufficiently similar. If there is sufficient clinical heterogeneity to expect that the underlying treatment effects differ between trials, or if substantial statistical heterogeneity is detected, we will use a random-effects analysis to produce an overall summary when an average treatment effect across trials is considered to be clinically meaningful. The random-effects summary will be treated as the average range of possible treatment effects, and we will discuss the clinical implications of treatment effects differing between trials. If the average treatment effect is not clinically meaningful, we will not combine trials. If we use random-effects analyses, the results will be presented as the average treatment effect with 95% CIs, and the estimates of T2 and I2 (Deeks 2001).

Subgroup analysis and investigation of heterogeneity

We will perform subgroup analysis to compare the following factors:

  1. baseline emergency and post-emergency situations where the intervention was launched;

  2. the context - whether the setting was urban or rural;

  3. inclusion and exclusion of another component of community-based approach, e.g. 'community mobilisation, 'supplementary feeding programme' and 'stabilisation centres';

  4. baseline HIV status of the children included where the intervention was launched;

  5. difference in the regions (middle- versus low-income countries, or sub-Saharan Africa, Middle East, North Africa, South Asia, East Asia, Pacific, Latin America and Caribbean countries).

Therefore,the main focus of this review will be to explore possible heterogeneity due to the above mentioned variables using meta-regression analysis if feasible. If sufficient studies are not identified, we will explore heterogeneity via different techniques, either visually via bubble plots or via box plots (displaying medians and ranges). Each of the modifiers will be evaluated in subgroup analysis and interaction tests will be considered. We will use the Comprehensive Meta analysis software for meta-regression where required (Comprehensive Meta Analysis 2005).

Sensitivity analysis

We will perform a random-effects analysis because we believe there will be heterogeneity among the studies in terms of populations, interventions, comparators, outcomes and settings. We believe it is very important to consider that selected studies may be different to some extent in terms of context. We will explore whether the effect estimator changes when using a fixed-effect model in a sensitivity analysis. Furthermore, we will conduct sensitivity analysis to assess to what extent the presence of adequate sequence generation and allocation concealment affects the results across studies.


Appendix 1. MEDLINE (Ovid) search strategy

1. Malnutrition/
2. Starvation/
3. Infant nutrition Disorders/
4. Child Nutrition Disorders/
5. Wasting Syndrome/
6. ((acute* or extreme* or sever* or serious* or fatal*) adj (malnutrition or mal nutrition or malnourish* or mal nourish*)).ti,ab.
7. ((acute* or extreme* or sever* or serious* or fatal*) adj ((deficient* or deficienc* or disorder*) adj2 (nutrition* or nourish*))).ti,ab.
8. ((acute* or extreme* or sever* or serious* or fatal*) adj (underfeeding or under feeding or underfed or under fed or undernourish* or under nourish* or undernutrition or under nutrition)).ti,ab.
9. ((acute* or extreme* or sever* or serious* or fatal*) adj3 (starvation or starving or famine?)).ti,ab.
10. ((acute* or extreme* or sever* or serious* or fatal*) adj (wasting or weight loss)).ti,ab.
11. or/1-10 [Malnutrition]
12. Developing,kf.
13. (Africa or Asia or Caribbean or West Indies or South America or Latin America or Central America).hw,kf,ti,ab,cp.
14. (Afghanistan or Albania or Algeria or Angola or Antigua or Barbuda or Argentina or Armenia or Armenian or Aruba or Azerbaijan or Bahrain or Bangladesh or Barbados or Benin or Byelarus or Byelorussian or Belarus or Belorussian or Belorussia or Belize or Bhutan or Bolivia or Bosnia or Herzegovina or Hercegovina or Botswana or Brazil or Brasil or Bulgaria or Burkina Faso or Burkina Fasso or Upper Volta or Burundi or Urundi or Cambodia or Khmer Republic or Kampuchea or Cameroon or Cameroons or Cameron or Camerons or Cape Verde or Central African Republic or Chad or Chile or China or Colombia or Comoros or Comoro Islands or Comores or Mayotte or Congo or Zaire or Costa Rica or Cote d'Ivoire or Ivory Coast or Croatia or Cuba or Cyprus or Czechoslovakia or Czech Republic or Slovakia or Slovak Republic or Djibouti or French Somaliland or Dominica or Dominican Republic or East Timor or East Timur or Timor Leste or Ecuador or Egypt or United Arab Republic or El Salvador or Eritrea or Estonia or Ethiopia or Fiji or Gabon or Gabonese Republic or Gambia or Gaza or Georgia Republic or Georgian Republic or Ghana or Gold Coast or Greece or Grenada or Guatemala or Guinea or Guam or Guiana or Guyana or Haiti or Honduras or Hungary or India or Maldives or Indonesia or Iran or Iraq or Isle of Man or Jamaica or Jordan or Kazakhstan or Kazakh or Kenya or Kiribati or Korea or Kosovo or Kyrgyzstan or Kirghizia or Kyrgyz Republic or Kirghiz or Kirgizstan or Lao PDR or Laos or Latvia or Lebanon or Lesotho or Basutoland or Liberia or Libya or Lithuania or Macedonia or Madagascar or Malagasy Republic or Malaysia or Malaya or Malay or Sabah or Sarawak or Malawi or Nyasaland or Mali or Malta or Marshall Islands or Mauritania or Mauritius or Agalega Islands or Mexico or Micronesia or Middle East or Moldova or Moldovia or Moldovian or Mongolia or Montenegro or Morocco or Ifni or Mozambique or Myanmar or Myanma or Burma or Namibia or Nepal or Netherlands Antilles or New Caledonia or Nicaragua or Niger or Nigeria or Northern Mariana Islands or Oman or Muscat or Pakistan or Palau or Palestine or Panama or Paraguay or Peru or Philippines or Philipines or Phillipines or Phillippines or Poland or Portugal or Puerto Rico or Romania or Rumania or Roumania or Russia or Russian or Rwanda or Ruanda or Saint Kitts or St Kitts or Nevis or Saint Lucia or St Lucia or Saint Vincent or St Vincent or Grenadines or Samoa or Samoan Islands or Navigator Island or Navigator Islands or Sao Tome or Saudi Arabia or Senegal or Serbia or Montenegro or Seychelles or Sierra Leone or Slovenia or Sri Lanka or Ceylon or Solomon Islands or Somalia or Sudan or Suriname or Surinam or Swaziland or Syria or Tajikistan or Tadzhikistan or Tadjikistan or Tadzhik or Tanzania or Thailand or Togo or Togolese Republic or Tonga or Trinidad or Tobago or Tunisia or Turkey or Turkmenistan or Turkmen or Uganda or Ukraine or Uruguay or USSR or Soviet Union or Union of Soviet Socialist Republics or Uzbekistan or Uzbek or Vanuatu or New Hebrides or Venezuela or Vietnam or Viet Nam or West Bank or Yemen or Yugoslavia or Zambia or Zimbabwe or Rhodesia).hw,kf,ti,ab,cp.
15. ((developing or less* developed or under developed or underdeveloped or middle income or low* income or underserved or under served or deprived or poor*) adj (countr* or nation? or population? or world)).ti,ab.
16. ((developing or less* developed or under developed or underdeveloped or middle income or low* income) adj (economy or economies)).ti,ab.
17. (low* adj (gdp or gnp or gross domestic or gross national)).ti,ab.
18. (low adj3 middle adj3 countr*).ti,ab.
19. (lmic or lmics or third world or lami countr*).ti,ab.
20. transitional countr*.ti,ab.
21. or/12-20 [LMIC]
22. randomized controlled
23. controlled clinical
24. multicenter
25. (randomis* or randomiz* or randomly allocat* or random allocat*).ti,ab.
26. groups.ab.
27. (trial or multicenter or multi center or multicentre or multi centre).ti.
28. (intervention* or controlled or control group or compare or compared or (before adj5 after) or (pre adj5 post) or pretest or pre test or posttest or post test or quasiexperiment* or quasi experiment* or evaluat* or effect or impact or time series or time point? or repeated measur*).ti,ab.
29. or/22-28
30. exp Animals/
31. Humans/
32. 30 not (30 and 31)
34. meta
38. cochrane database of systematic reviews.jn.
39. comment
40. (systematic review or literature review).ti.
41. or/32-40
42. 29 not 41 [Methods filter]
43. 11 and 21 and 42

Contributions of authors

Yasir Shafiq, Dr Ali Faisal Saleem, Zohra Lassi and Dr Anita Zaidi drafted the protocol

Declarations of interest

  1. Yasir Shafiq: Not known

  2. Dr Ali Faisal Saleem: Not known

  3. Zohra Lassi: Not known

  4. Dr Anita Zaidi: Not known

Sources of support

Internal sources

  • None, Not specified.

External sources

  • None, Not specified.