Measurement of benefits in economic evaluations of nutrition interventions in low‐ and middle‐income countries: A systematic review

Abstract Economic evaluation of nutrition interventions that compares the costs to benefits is essential to priority‐setting. However, there are unique challenges to synthesizing the findings of multi‐sectoral nutrition interventions due to the diversity of potential benefits and the methodological differences among sectors in measuring them. This systematic review summarises literature on the interventions, sectors, benefit terminology and benefit types included in cost‐effectiveness, cost‐utility and benefit‐cost analyses (CEA, CUA and BCA, respectively) of nutrition interventions in low‐ and middle‐income countries. A systematic search of five databases published from January 2010 to September 2019 with expert consultation yielded 2794 studies, of which 93 met all inclusion criteria. Eighty‐seven per cent of the included studies included interventions delivered from only one sector, with almost half from the health sector (43%), followed by food/agriculture (27%), water, sanitation and hygiene (WASH) (10%), and social protection (8%). Only 9% of studies assessed programmes involving more than one sector (health, food/agriculture, social protection and/or WASH). Eighty‐one per cent of studies used more than one term to refer to intervention benefits. The included studies calculated 128 economic evaluation ratios (57 CEAs, 39 CUAs and 32 BCAs), and the benefits they included varied by sector. Nearly 60% measured a single benefit category, most frequently nutritional status improvements; other health benefits, cognitive/education gains, dietary diversity, food security, knowledge/attitudes/practices and income were included in less than 10% of all ratios. Additional economic evaluation of non‐health and multi‐sector interventions, and incorporation of benefits beyond nutritional improvements (including cost savings) in future economic evaluations is recommended.


| INTRODUCTION
Malnutrition is widely acknowledged by governments, international agencies, donors and researchers as a problem with diverse causes, requiring multiple strategies and the engagement of multiple sectors.
Sixty-one country governments are part of the United Nations (UN) Network for Scaling Up Nutrition (SUN) Movement, which explicitly calls for a multi-stakeholder and multi-sectoral approach to improving nutrition outcomes (Scaling Up Nutrition, 2020). This approach combines both nutrition-specific interventions (those that address the immediate causes of malnutrition, such as micronutrient supplementation, treatment of acute malnutrition and promotion of appropriate dietary and feeding behaviours) and nutrition-sensitive interventions (those that address the underlying causes, such as ensuring child protection, women's empowerment, agricultural production and adequate water and sanitation; Ruel et al., 2013). To assist countries in developing multi-sectoral nutrition strategies, SUN and the UN Renewed Efforts Against Child Hunger (REACH) initiative published the Compendium of Actions for Nutrition, a wide-ranging menu of nutrition-specific and nutrition-sensitive interventions (World Food Programme, 2016).
Given the diversity of nutrition-specific and nutrition-sensitive interventions, robust economic evaluations of multi-sectoral approaches are essential for setting priorities and efficiently allocating resources, particularly in low-and middle-income countries (LMICs) that bear the disproportionate burden of malnutrition. Compared to other global health conditions, such as HIV/AIDS, malaria, other infectious diseases and non-communicable diseases, the costeffectiveness evidence base for nutrition interventions is limited.
Notably, a bibliometric review of 614 economic evaluations of health interventions in LMICs found that only 3% pertained to malnutrition and/or anaemia (Pitt et al., 2016). In addition, less than 6% of the Global Health CEA registry, a database of cost-effectiveness studies evaluating a range of health interventions worldwide, cover interventions to address nutritional deficiencies (Center for Evaluation of Value and Risk in Health [CEVR], 2019). With the growing awareness of the importance of economic evaluation evidence for resource allocation, priority setting, scaling of effective solutions and global and national funding decisions, recently there has been a flurry of systematic reviews that shed light on the costs and benefits of interventions to address nutritional deficiencies in global settings. Two recently published systematic reviews of economic evaluations of interventions provide evidence on preventive nutrition interventions, such as supplementation, infant and young child feeding, therapeutic nutrition interventions (interventions to treat undernutrition and micronutrient deficiencies), fortification and cash transfers linked to improved nutritional outcomes (Njuguna et al., 2020;Ramponi et al., 2020). A third recent study by Das et al. (2020) provides a systematic review of both the effectiveness and cost-effectiveness of interventions that manage acute malnutrition in children in LMICs. Baek et al. (2021)  It may not be coincidental that there are persistent evidence gaps for nutrition-sensitive interventions and multi-sectoral approaches. First, it is a substantive challenge to capture and value the diverse benefits associated with multi-sectoral strategies to improve health outcomes. Multi-sectoral nutrition strategies produce a wide variety of tangible and intangible benefits to individuals, households and communities. Many nutrition-sensitive interventions have been shown to significantly improve dietary practices, enhance care practices and reduce the prevalence of disease (Sharma et al., 2021). These interventions lead to improved nutrition and health outcomes through food production, nutrition-related knowledge, agricultural income and women's empowerment (Sharma et al., 2021). Tangible outcomes can readily be presented in monetary terms and included in economic analysis. These include monetary outcomes like changes in food production, agricultural income and labour productivity. They also include health and nutrition outcomes such as stunting and wasting which have associated morbidity and mortality that can be valued for economic purposes. On the other hand, intangible outcomes, such as women's empowerment, are often measured using methods that are more difficult to value, such as qualitative inquiry or the use of indices.
Health economic evaluation is concerned with the health and monetary benefits resulting from a policy or intervention. Health benefits can be measured with a variety of health and nutrition metrics and can be assigned monetary values. Monetary benefits may refer to averted medical costs or increases in productivity from an intervention. There are three main types of economic evaluation

Key messages
• Current economic evaluations often underestimate the total sum of benefits that can arise from nutrition interventions. Comprehensive benefit measurement of some nutrition programmes may require further methodological research.
• In the near-term, economic evaluations of multi-sectoral nutrition interventions should include potential cost savings from improved nutrition in their calculations and assess the potential for benefits unrelated to nutrition. If the range of benefits is diverse and can be monetised, benefit-cost analysis may be the preferred evaluation method.
• Economic evaluations of nutrition-sensitive interventions from agriculture, water, sanitation and hygiene (WASH), and gender empowerment sectors, are needed to fill an evidence gap on costs and benefits of multisectoral approaches to improved maternal and child health and nutrition.
comparing the costs with the consequences of an intervention: costeffectiveness analysis (CEA), cost-utility analysis (CUA) and benefitcost analysis (BCA; Drummond et al., 2015). Economic evaluations require distinct considerations when used to evaluate nutrition strategies. CEA compares costs to one specific outcome at a time (such as cost per case of wasting averted) in a cost-effectiveness ratio. Therefore, these ratios cannot capture the full range of benefits resulting from a multi-sectoral intervention. CUA calculates costs in terms of health-adjusted life years such as disability-or qualityadjusted life years (DALYs or QALYs). These measurements express various health-related outcomes in terms of 'utility', an economic concept related to the level of 'satisfaction' (or lack thereof) experienced in various health states. Utility-based metrics can facilitate comparisons between interventions addressing different diseases, and they enable analysts to include multiple health states in one costutility ratio; however, the health benefits included in the evaluation are still limited to death or disability. Finally, a BCA presents all intervention benefits in monetary terms, and therefore each analysis can include a wider range of current and future health and economic benefits. However, valuation of intangible benefits in BCA studies is restricted to the available (and limited) evidence on willingness to pay and revealed preferences for health and nutrition outcomes. Despite the challenges involved in assigning monetary value to intangible impacts and outcomes, for the remainder of this paper, we use the term benefit to refer to all tangible and intangible impacts and outcomes of multi-sectoral nutrition interventions which have intrinsic value to individuals, households and communities in LMICs. Furthermore, the methodology and assumptions used for determining the monetary value of improved health can vary considerably between studies, and some decision-makers may object to the concept of translating health to a monetary value (Mills, 2014).
Conventional economic evaluations have typically considered a single health sector whose target is to maximise health or minimise costs. Remme et al. (2017) note that this approach fails to consider that multiple sectors contribute to population and individual health outcomes, and that many of the goods and services produced by the healthcare system have benefits beyond health. Nutrition experts have expressed concerns about using economic evaluation methods (such as cost-effectiveness) given the heterogeneity of nutrition programmes and the challenge of capturing nutrition benefits, especially when some intervention's primary objectives, such as increased food production, fall outside the health sector domain (Levinson & Herforth, 2013). While other assessment options may exist, based on effectiveness of increasing food production and food security and economic viability, if donors or governments must allocate scarce resources across competing sectoral demands, then improved cost-effectiveness and benefit-cost ratios are essential for comparing across investments and are a consideration for decision making. In their absence, it may be difficult to advocate for nutrition as it competes with other government priorities.
Compounding the issue is the difficulty in comparing findings from economic evaluations of multi-sectoral interventions that use different methodologies. Researchers and practitioners from different disciplines often use distinct terminology to describe comparable analytical approaches. More importantly, interventions-particularly those from different sectors-usually have different objectives and intended proximal and distal outcomes. The heterogeneity of multisectoral programmes and their study designs, the range of benefits measured and valued, and concerns related to quality assessment have all been noted by recent systematic reviews (Baek et al., 2021;Njuguna et al., 2020;Ramponi et al., 2020).
The overall aim of this systematic review is to describe the full range of benefits that have been included or excluded from the current literature on cost-effectiveness and benefit-cost of nutrition interventions. We have chosen to focus only on the benefits included in a study or as part of a cost-effectiveness or benefit-cost ratio, rather than on the specific intervention costs measured in the evaluations. The specific objectives of this systematic review are to: (1) characterise the types of nutritionspecific and nutrition-sensitive interventions included in recent economic evaluations and (2) assess the range of terminology and methodological approaches used to value the nutrition-related benefits of these interventions. We believe this can help to identify research gaps and improve the quality and design of future studies conducted by interdisciplinary teams of nutritionists, epidemiologists and economists. These findings will inform the design of future economic evaluations of multi-sectoral nutrition interventions seeking to capture and value the broadest possible range of health and economic benefits.

| METHODS
This systematic review complies with the 'preferred reporting items for systematic reviews and meta-analyses' (PRISMA) checklist for conducting a systematic review (Moher et al., 2009), with the exception of evaluation of bias since the aim of the study was to provide a qualitative assessment of benefits rather than to quantify the magnitude of those benefits. Our study protocol is detailed in Appendix A and summarised below. Given no human subjects were involved in this review, an institutional review board was not needed.

| Inclusion criteria
We searched for English-language, peer-reviewed, empirical evaluations of nutrition-related interventions, conducted in one or more LMICs, published from 1 January 2010 to 26 September 2019, and reporting at least one ratio comparing intervention costs and benefits (i.e., cost-effectiveness ratios for CEAs, cost-utility ratios for CUAs or benefit-cost ratios for BCAs; Gillespie & van den Bold, 2017). We and national policy actions that would not have an isolated impact on nutrition outcomes, and vertical global health interventions preventing a range of infectious diseases that typically fall outside of nutrition-specific interventions (such as vaccination and prevention of mother-to-child transmission of HIV) were excluded. In addition, some interventions were expanded beyond the Compendium, where nutrition experts have explicitly demonstrated their effectiveness for improving maternal or child nutrition Keats et al., 2021); for instance, all malaria prophylaxis and treatment interventions were included, whereas the Compendium only listed intermittent preventive treatment of malaria for pregnant women and distribution of bed nets. Table 1 lists the 78 compendium interventions that we considered. Twenty-three interventions were from the agriculture/food sector, 27 were from the health sector, 12 were from the water, sanitation and hygiene (WASH) sector, and 16 were from the social protection sector.
We also only included studies that measured at least one nutritionrelated outcome. This allowed us to narrow our focus to studies of interventions that included nutrition as a primary or secondary objective and were designed as such, as opposed to interventions that may have incidentally changed nutritional status. For instance, improving access to fertiliser and other agricultural inputs is a Compendium intervention, but if a study evaluating this intervention measured only changes in agricultural yield, it would have been excluded from our review. Nutritionrelated outcomes were defined as the following: improvement in nutritional status; monetary savings from averting a nutritional disorder; food security; dietary diversity; nutritional knowledge, attitudes and/or practices; diarrhoeal incidence; household income; and women's empowerment Herforth & Harris, 2013).

| Search strategy
We searched six databases for studies meeting our inclusion criteria: PubMed, Embase, Web of Science, EconLit, Cinahl and the Cochrane Central Register of Controlled Trials. We developed a list of search terms targeting these criteria. This search strategy included several terms for undernutrition (e.g., acute malnutrition and micronutrient deficiencies), since the vast majority of interventions mentioned in the Compendium are intended to address nutritional deficiencies.
These terms were then optimised for each database by an information specialist before running the search (the full search strategy is presented in Tables A1a-e). Endnote was used to identify duplicate results across different databases (Clarivate, 2020). An additional 24 articles were found through expert consultation and the World Health Organization e-Library of Evidence for Nutrition Actions (eLENA; World Health Organization, 2020).

| Screening and assessment
Two reviewers (Jolene Wun, Christopher Kemp or Devon Bushnell) used the Covidence tool (Covidence) to independently assess each study's title and abstract for inclusion and resolve discrepancies. One reviewer (Jolene Wun or Devon Bushnell) then reviewed the full text of screened studies, and if all inclusion criteria were determined to have been met, the reviewer proceeded to enter key study information in a structured data abstraction form. A second reviewer (Jolene Wun, Christopher Kemp, Chloe Puett or Devon Bushnell) then verified each abstraction and the two resolved any differences through discussion. Data were abstracted from included studies at two levels. At the study level, the abstraction form contained fields for study details (World Bank region, type of intervention[s] and terminology used to describe benefits). At the economic evaluation ratio (hereafter referred to as ratio) level, the abstraction form contained fields for the type of ratio included (CEA, CUA or BCA), and the type of benefit.
Benefit categories are defined in Table 2.

| Analyses
For the studies meeting the inclusion criteria, we calculated the number of studies by sector (food/agriculture, health, WASH, social protection or multi-sector). We also tabulated the number and percentage of ratios including each benefit type; and for ratios that included a nutrition status improvement, the specific illness averted. All analyses were conducted in Microsoft Excel.  Table A2.

| Study-level descriptive statistics
Studies in our sample covered 44 of the 78 selected Compendium nutrition interventions. Table 3 presents a breakdown of specific interventions evaluated. The most commonly assessed interventions were in the health sector: management of severe acute malnutrition/wasting (n = 12), zinc supplementation (n = 12) and oral rehydration for diarrhoea (n = 11). The most commonly assessed interventions in the food/agriculture sector were mass fortification (n = 9), and biofortification (n = 7). The greatest number of economic evaluations for social protection were for food vouchers (n = 4). The greatest number of economic evaluations in the WASH sector were for household water treatment and storage (n = 5).
Almost half of the studies (41 studies, or 44%) were from Sub-Saharan Africa; 22% were from South Asia, 15% were from East Asia & the Pacific, 11% from multiple world regions, and the rest in one of the other regions (Europe and Central Asia, Latin America of studies), followed by 'outcome' (62%), and 'impact' (52%). However, the majority of studies (80%) used more than one term, with 27% using at least four. Agriculture sector studies most often referred to 'benefit' (76%), followed by 'impact', then 'outcome'. Only 32% of agriculture studies referred to 'effect' and 'savings'. Health sector studies primarily referred to 'effect' (73%), followed by 'outcome', 'benefit' and 'impact'. Only 25% of health sector studies referred to 'savings'. For social protection studies, the terminology most commonly used was 'outcome' (86%), followed by 'effect' and 'impact' (71% each); 'benefit' and 'savings' were found in 13% of the studies. WASH studies predominantly referred to 'benefit' (78%), followed by 'impact' and 'savings' (56% each).
Economic evaluation methods appeared to be related to the use of specific terminology. For example, 97% of studies including a BCA ratio referred to 'benefit,' while CEAs and CUAs generally referred to 'outcome' and 'effect'. Notably, the term 'impact' was used with similar frequency in studies with CUAs (61%) and BCAs (63%) but much less frequently in studies calculating CEAs (33%).

| Ratio-level descriptive statistics
The 93 included studies estimated a total of 128 economic evaluation ratios. Of these ratios, 57 were CEAs (44%), 39 were CUAs (30%) and 32 BCAs (25%). Of the 128 ratios analysed, 76 (59%) measured a single benefit category, 37 (28%) measured two benefit categories, and the remaining 15 (12%) included three or more benefit categories. all six of these ratios came from one study assessing various attitudes and practices. Increases in household income or assets were included in a higher number of ratios assessing food/agriculture interventions (9 out of 27 ratios, or 33%), compared to 13% of ratios assessing social protection interventions, 6% of evaluations assessing multiple sectors and none of the ratios studying health or WASH interventions.
Notably, only around one-third of all ratios considered cost savings of any kind. For instance, 47% of BCA ratios included beneficiary cost savings, compared to 12% of CEAs and 13% of CUAs.
Cost savings by beneficiaries were more commonly included (22%) than cost savings by the provider (13%), and of those costs, direct costs were more commonly included (18%) as compared to indirect costs (14%).
Within the category of nutritional status improvements, Figure 2 further disaggregates the specific anthropometric, mortality and morbidity benefits measured in CEAs, and nutrition-related DALYs  To move towards standard measurement across a range of multi-sectoral interventions, standardised language around economic evaluations is needed. We assessed the range of terminology and methodological approaches employed to value benefits when compared against intervention costs. We also identified distinctions in terminological use across sectors and study types. There is room for standardising the terminology used in economic evaluations of multi-sectoral nutrition approaches and interventions. Ideally, terminology could be standardised for use in economic evaluation, depending on the type of analysis to be conducted (CEA, CUA and BCA) and type of benefit that is being assessed (e.g., monetary gain is a benefit, nutrition status improvement is an outcome or effect). In practice, it will take time and coordinated effort to obtain consensus across sectors and disciplines, since the inconsistency across studies reflects differences in both impact evaluation methods and focus across both sectors and disciplines. For example, nutritionists conduct evaluations to measure impact or effectiveness of an intervention, and may focus more on nutritional status, caring practices, educational outcomes, food security or diet-related changes. Agriculturalists measure impact in crop yields and net incomes, but rarely capture impacts on health or nutrition outcomes. Economists then use the available information on impact or effectiveness results to value the full range of current or future health and economic benefits, converting them to monetary values or utility-based measurements when feasible.
Economic evaluation of nutrition interventions is challenged by the breadth of outcomes that they affect. In addition to numerous measurable nutrition and health outcomes, there are benefits related to agricultural productivity, income generation, food security, dietary diversity and women's empowerment. This study identifies which benefits have been captured in studies to date and the differences across sectors. The choice of which effects and benefits to include in economic evaluation ratios and the type of economic evaluation selected was found to be strongly related to the sector of the intervention. Health sector and multisectoral evaluations tended to focus on nutritional status improvements and conduct CEAs and CUAs. Other sectors were more mixed in which benefits to include and, with the exception of social protection, conducted BCAs more often.  (27) Health (54) Social protection (16) WASH (14) Multiple (17) CEA (57) CUA (39) BCA (    F I G U R E 2 Distribution of conditions included in nutrition-related improvement in economic evaluation ratios, for four categories of improvements. 'Other' includes iron deficiency anaemia (3%), hepatocellular carcinoma (3%), helminth infection (2%) and folate deficiency (2%) Comparing the list of potential benefits from multi-sectoral nutrition interventions (Table 2) with the array of benefits found in this systematic review highlights the predominance of some conditions within the current evidence base-namely wasting, stunting, diarrhoea, anaemia and vitamin A deficiency (as outlined in Figure 2)-and the omission of women's empowerment and mental/social benefits regardless of sector. In addition, cognitive improvements, dietary diversity, food security and changes in knowledge, attitude and practices were measured, but rarely; and cognitive improvements were frequently measured as productivity gains (e.g., higher wages from increased school attendance resulting from improvements in nutrition). These calculations may be highly sensitive to assumptions about future labour markets and economic prospects. Some of the gaps in the existing evidence are due to practical or methodological challenges in benefit measurement, such as placing a monetary value on benefits that do not have a market value.
These are intractable challenges that will require future research to advance methods for measurement and quantification.
However, other gaps in counting benefits are easier to address in the shorter-term. For example, only one-third of studies included cost savings in terms of averted health care and/or time spent seeking health care.
Cost savings can be included in any of the three types of economic evaluation (CEA, CUA or BCA) and should be included more often in economic evaluations for nutrition interventions. Additionally, the majority of economic evaluations in this review (59%) included just one benefit, and about a quarter included two. With the exception of WASH sector evaluations, the inclusion of benefits unrelated to nutrition was relatively rare, so other sectors could consider expanding the range of benefits beyond nutrition and conduct BCAs rather than CUAs or CEAs if their benefits are diverse and can be monetised. There also is scope for more studies to include cognitive, education, or productivity gains associated with investments in nutrition.
This review had several limitations. First, we did not include all Compendium interventions. For example, we included malaria treatment and interventions, given their recognised effectiveness in preventing maternal and child nutrition, but excluded other interventions that prevent infant and childhood diseases, notably immunisations and prevention of mother-to-child transmission of HIV, which may bias our findings. We also excluded studies that did not explicitly assess a nutrition-related outcome. We therefore did not evaluate interventions that improve population nutrition incidentally. Second, studies conducted in high-income countries were excluded, though they may represent a significant proportion of nutrition-related economic evaluations. This review focused on the unique challenge of implementing and evaluating complex nutrition programmes in low-resource settings. Third, this review was focused on the science of economic evaluation, and we excluded unpublished and nonpeerreviewed studies. Our results may be biased towards investigators from high-income, English-speaking settings given barriers to academic publication in English among investigators from lower-income settings. Finally, our search strategy included explicit search terms for undernutrition and not overnutrition. This review thus may not reflect the full breadth of economic research on strategies to combat nutrition-related non-communicable diseases (Nugent et al., 2020).

ACKNOWLEDGMENTS
The authors appreciate the input provided by Amy Margolies in the development of this manuscript. The authors also would like to thank the Bill and Melinda Gates Foundation for financial support of this study.

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS
CL, CK and JW designed the research study. JW, CK, CP and DB analysed the data. JW, CK, CP, JC and CL wrote the paper.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. • Nutrition disorders (micronutrient deficiencies, anaemia, stunting, wasting, acute malnutrition and protein-energy malnutrition) • Disability-adjusted life years or mortality associated with nutrition disorders • Any other benefits/savings from a reduction in a nutrition-related disease/disorder (including averted treatment costs and improved productivity) Note: Measures of effect/benefit must be included in the economic analysis (e.g., part of an incremental cost-effectiveness ratio, benefit-cost ratio or net benefit estimate).

Setting: Any low and middle-income country (LMIC) per World
Bank criteria.
Type of study: Any cost-effectiveness, cost-utility, or costbenefit analysis of any study design (e.g., pre-post, quasiexperimental, RCT and modelled).
Dates: On or after 1 January 2010

Languages: English only
Types of articles: Original research articles in peer-reviewed journals (e.g., no reviews, conference abstracts, book chapters or other grey literature reports).

Study selection process
Extract articles b. Export to Google Sheets 11. Complete data extraction forms for articles to be included: a. Two reviewers will be assigned to each article. One will abstract data for included article on the data abstraction form.
The other will validate data abstraction and note any discrepancies.
12. All discrepancies resolved through discussion with the reviewer team. T A B L E A1b Search strategy for Embase (PubMed) #1 Search "cost effect*" or "cost-effect*" or "cost benefit" or "cost-benefit" or "cost utility" or "cost-utility" or "return on investment" Field: Title/ Abstract Search "cost analysis" or costs or cost or economics or expenditures or "economic evaluation*" or "out of pocket" or expenses Field: Title/ Abstract #4 Search ((#1) OR #2) OR #3 #5 Search ("acute malnutrition" or malnutrition or "nutritional care" or "lactating women" or "child nutrition" or "infant nutrition" or "maternal nutrition" or undernutrition or "under-nutrition" or "severe acute malnutrition" or SAM or CMAM or "community management acute malnutrition" or wasting or wasted or malnourish* or "acutely malnourished" or marasmus or OR GAM OR MAM OR wasting OR wasted OR malnourish* OR â€oeacutely malnourished" OR marasmus or income or "diet* diversity" or consumption or intake or food security or "wom* empowerment" or "female empowerment" or diarrhoea or malaria or measles or pneumonia or meningitis or anaemia or anaemia or deficiency or stunt* Field: Title/Abstract Search "Puerto Rico" or Romania or Rumania or Roumania 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 Uzbekistan or Uzbek or Vanuatu or "New Hebrides" or Venezuela or Vietnam or "Viet Nam" or "West Bank" or Yemen or Zambia or Zimbabwe Field: Title/Abstract #17 Search ((#16) OR #15) OR #14 #18 Search (#17) AND #13 #19 Search (Protein* or "multiple micronutrient*" or calcium or "folic acid" or iron or vitamin* or zinc or "fatty acid*") and supplement* Field: Title/ Abstract #20 Search biofortification Field: Title/Abstract #21 Search ALPHA or "antenatal psychosocial health assessment" or "postpartum depression" or "delayed cord clamping" or "paid maternity leave" or "Preterm massage" or Breastfeed* or "complementary feeding" or "infant feeding" or "home fortification" or "birth control" or "birth spacing" or "delay* pregnancy" Field: Title/Abstract #22 Search (malaria or deworm* or diarrhoea or "air pollution") and (control or prevent* or treatment) Field: Title/Abstract #23 Search "ready to use therapeutic food" OR "ready to use food" OR "ready to use supplementary food" OR "plumpy nut" OR "plumpynut" OR plumpysup OR imunut OR plumpy* OR nutributter OR FBF OR "fortified blended flour" OR "super cereal" Field: Title/Abstract #24 Search "kitchen garden*or community garden*" or "home garden*" or "school garden*" or crops or horticulture or acquaculture Field: Title/ Abstract #25 Search "food safety" or "aflatoxin prevention" Field: Title/Abstract #26 Search "cash crop*" Field: Title/Abstract #27 Search livestock or cattle or poultry or "dairy farm*" or "animal husbandry" or "animal rearing" or fish* or meat or chicken or goat* or cow* or cattle or pig* or sheep* or fish Field: Title/Abstract Search "faecal sludge management" or "faecal waste management" or "child faeces disposal" Field: Title/Abstract #39 Search (sanitation or latrine or toilet) and marketing Field: Title/Abstract #40 Search "drinking water" and (chlorine or filter or treatment or storage or improv* or safe) Field: Title/Abstract #41 Search ("source water" or "water supply") and improv* Field: Title/Abstract #42 Search voucher* or "school meal*" or "cash transfer*" or "In-kind transfer*" or "health insurance" or insurance Field: Title/Abstract T A B L E A1c Search strategy for Web of Science. (Indexes = SCI-EXPANDED, SSCI, A&HCI, CPCI-S, CPCI-SSH, BKCI-S, BKCI-SSH, ESCI, CCR-EXPANDED) Refined by: PUBLICATION YEARS: OR 2010OR 2013OR 2012 1 TOPIC: ("cost effect" or "cost benefit" or "cost benefit analysis") ORTOPIC: (("economic evaluation*" or "out of pocket")) 2 TOPIC: (("acute malnutrition" or malnutrition or "nutritional care" or "lactating women" or "child nutrition" or "infant nutrition" or "maternal nutrition" or undernutrition or "under-nutrition" or "severe acute malnutrition" or SAM or CMAM or "community management acute malnutrition" or wasting or wasted or malnourish* or "acutely malnourished" or marasmus)) ORTOPIC: (("diet* diversity" or consumption or intake or food security or "wom* empowerment" or "female empowerment" or diarrhoea or malaria or measles or pneumonia or meningitis or anaemia or anaemia or deficiency or stunt*)) ORTOPIC: (DALY or QALY or "nutrition disorder*" or "growth disorder*" or Kwashiorkor) 3 #2 AND #1 4 TOPIC: (("developing country" or "developing countries" or "low and middle income countries" or LMIC or asia or africa or "south america" or oceania or "latin america" or caribbean)) ORTOPIC: (Afghanistan or Albania or Algeria or Angola or Antigua or Barbuda or Argentina or Armenia 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 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") ORTOPIC: (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) 5 TOPIC: (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 Ethiopia or Fiji or Gabon or "Gabonese Republic" or Gambia or Gaza or Georgia or Georgian 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 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 Lebanon or Lesotho or Basutoland or Liberia or Libya 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) ORTOPIC: ((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 "Puerto Rico" or Romania or Rumania or Roumania 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")) ORTOPIC: (("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 Uzbekistan or Uzbek or Vanuatu or "New Hebrides" or Venezuela or Vietnam or "Viet Nam" or "West Bank" or Yemen or Zambia or Zimbabwe)) 6 # 5O R# 4 7 #6 AND #3 8 TOPIC: ("diet supplementation" or "nutritional supplement*" or biofortification) ORTOPIC: ("antenatal psychosocial health assessment" or ("postpartum depression" or "delayed cord clamping" or "paid maternity leave" or "Preterm massage" or Breastfeed* or "birth control" or "birth spacing" or "delayed pregnancy")) 9 TOPIC: (((malaria or deworm* or diarrhoea or "air pollution") and (control or prevent* or treatment))) ORTOPIC: (("ready to use therapeutic food" or "ready to use supplementary food" or "plumpy nut" or plumpysoy or imunut or plumpy* or nutributter or FBF or "fortified flour" or "super cereal").) ORTOPIC: (("kitchen garden" or "community garden" or "home garden" or "school garden" or crops or horticulture or acquaculture) or ("food safety" or "aflatoxin prevention" or "cash crops") 10 TOPIC: ((livestock or cattle or poultry or "dairy farm*" or "animal husbandry" or "animal rearing" or fish* or meat or chicken or goat* or cow* or cattle or pig* or sheep* or fish)) ORTOPIC: (Irrigation or biodiversity) ORTOPIC: ("agricultural education" or ("crop rotat*" or intercrop* or "insect farm" or "food storage")) 11 TOPIC: ((Malt* or dry* or pickl* or cur* or preserv*) and (food or vegetable or fruit)) ORTOPIC: (Food and (tax or subsid* or regulation or marketing)) ORTOPIC: (hygiene or handwash* or sanitation or "drinking water" or "water supply") 12 TOPIC: ((voucher* or "school meal*" or "cash transfer*" or "In-kind transfer*" or "health insurance")) 13 #12 OR #11 OR #10 OR #9 OR #8

#13 AND #7
T A B L E A1d Search strategy for Cinahl (EBSCOhost) and EconLIT (EBSCOHost)