Impact of cash transfer programs on birth and child growth outcomes: systematic review

Abstract To investigate the impact of cash transfer (CTs) on birth outcomes, including birth weight, low birth weight and prematurity, as well as child physical growth were included, as assessed by anthropometric indices in children under five years of age. Searching was performed using the PubMed/Medline, Embase, LILACS, Cochrane Library, Scopus and Web of Science databases. Quantitative observational, experimental and quasi-experimental. Eleven studies were included in the review. The majority (81.8%) were carried out in low-and middle-income countries and most involved conditional CTs (63.6%). Four were clinical trials and seven were observational studies. Conditional CTs were found to be associated with a reduction in height-for-age (-0.14; 95%CI -0.27, -0.02); (OR 0.85; 95%CI 0.77-0.94); (OR = 0.44; 95%CI 0.19-0.98), a significantly reduced chance of low weight-for-age (OR = 0.16; 95%CI -0.11-0.43), low weight-for-height (OR = -0.68; 95%CI -1.14, -0.21), and low weight-for-age (OR = 0.27; 95%CI 0.10; 0.71). Unconditional CTs were associated with reduced birth weight (RR = 0.71; 95%CI 0.63-0.81; p < 0.0001) and preterm births (RR = 0.76; 95%CI 0.69-0.84; p < 0.0001). Conditional CTs can positively influence birth outcomes and child growth.


Introduction
An estimated 90-117 million children live in poverty worldwide 1 , a condition which reduces a family's capability to provide children with the care and attention necessary to ensure adequate growth and development in the first five years of life.Poverty is considered a social determinant of health with multidimensional consequences 2,3 .Previous studies have highlighted relationships between poverty and increases in infectious and parasitic diseases, protein-calorie malnutrition and micronutrient deficiencies, as well as higher rates of hospitalization and death among children [3][4][5] .
Social protection policies are important interventions to reduce poverty and protect nutritional status and health of children and newborns, especially considering the strength of relationships between poverty and negative nutritional outcomes, e.g., low birth weight, premature birth and delayed growth 3,6,7 .
In this context, cash transfer programs (CT) have been implemented in several countries, especially those considered as low-or middle-income.A form of public policy aimed at reducing poverty and social inequality, cash transfers provide a source of monthly income to previously registered eligible beneficiaries.Among the main advantages of CTs are the improved well-being of families, income redistribution and the promotion of social inclusion 8,9 .
CTs can be classified according to the presence or absence of eligibility conditions.In unconditional income transfer programs (UCT), monetary transfer occurs with no action required from beneficiaries 19 , while in conditional income transfer programs (CCT), monthly benefits are linked to the fulfillment of specific education and/or health stipulations 13,20 .
To date, no systematic reviews have attempted to evaluate the effectiveness of these strategies on child health and nutrition in different economic contexts.Thus, in light of the relevance of providing consistent evidence on the impact of CTs on child health and nutrition, while also considering CT type (conditional/unconditional), here we endeavored to systematically analyze studies evaluating the effects of these programs on prematurity, low birth weight and other indicators of physical growth among children aged five years or less.Knowledge on these effects serves as a strategic tool for public policymakers to administer social programs aimed at ensuring the healthy development of babies from birth through infancy.

Methods
The present review was developed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement 21 , and has been registered with the International Prospective Register of Systematic Reviews (PROSPERO): CRD42021255570

search strategy
Searching was performed on the PubMed/ Medline, Embase, LILACS (Virtual Health Library), Cochrane Library, Scopus and Web of Science electronic databases.Additionally, grey literature was consulted by an expert on the topic of this review (Figure 1).
No language, geographic or publication date restrictions were applied.Results were obtained through July 2021 and update until November 8th, 2022.The keywords employed when performing searches were identified using Medical Subject Headings (MESH) vocabulary and adapted to each database using the Health Sciences Descriptors (DeCS) thesaurus and Embase Subject Headings (Emtree) (Chart 1).

Participants, exposure, comparisons and outcomes
To answer the guiding question: "What are the impacts of cash transfer programs on birth outcomes and physical growth in children aged under five years?", the acronym PICO (population/problem, intervention/exposure, comparison, outcome) was developed jointly with specialists in the field, as delineated in Chart 2. The exposure/intervention of interest was conditional and/or unconditional income transfer programs targeting socioeconomically vulnerable families or individuals.Children of families who were not beneficiaries of CTs were considered as controls.
The following outcomes were considered: The nutritional status of children under five years, as assessed by the anthropometric measures weight-for-height (W/H), weight-for-age (W/A), heightfor-age (H/A) and body mass-for-age (BMI/A) 22 ; birth weight and low birth weight (< 2,500 g); preterm birth, defined as birth occurring before the 37 th week of pregnancy (< 259 days, or 36 weeks and 6 days) 23 (Chart 2).
Due to substantial heterogeneity among the obtained results, a narrative synthesis was adopted to present our findings (Charts 3 and 4).

eligibility criteria
Quantitative observational and experimental studies, as well as quasi-experimental published articles, were included regardless of country income classification (low, middle or high income).
The present systematic review excluded qualitative research, books or chapters of narrative and scientific books, editorials, opinion articles, literature reviews (narrative, integrative, and systematic, with or without meta-analysis, scoping review/rapid review), studies in which food and/ or nutritional supplements were offered, as well as those investigating food vouchers or emergency cash transfers.

study selection and data extraction
All articles identified in the searched databases were entered into the Rayyan application 24 to assist in screening.Four reviewers determined eligibility through the analysis of titles and abstracts.Discrepancies between reviewers were resolved by discussion and in collaboration with a fifth reviewer.The reasons for exclusion of all full-text articles are detailed in Figure 1.
Full texts were reviewed using a standardized form to determine final inclusion in the present review.A data extraction template enabled the collection of information on first author, year of publication, manuscript title, study location, population (study size and description of groups), design, exposure variables (type of CT, monthly transfer value, presence/absence of con-Chart 1. search strategy employed for PubMed (by Medline), EMBASE, Lilacs (by Virtual Health Library), Cochrane Library, Scopus and Web of Science (2021).#1 "Infant" [Mesh Terms] OR "Infants" OR "Infant, Newborn" [Mesh Terms] OR "Infants, Newborn" OR "Newborn Infant" OR "Newborn Infants" OR "Newborns" OR "Newborn" OR "Neonate" OR "Neonates" OR "Child, Preschool" [Mesh Terms] OR "Preschool Child" OR "Children, Preschool" OR "Preschool Children" OR "Child" [Mesh Terms] OR "Children" #2 "Public Policy" [Mesh Terms] OR "Policies, Public" OR "Policy, Public" OR "Public Policies" OR "Social Protection" OR "Protection, Social" OR "Population Policy" OR "Policies, Population" OR "Policy, Population" OR "Population Policies" OR "Social Policy" OR "Policies, Social" OR "Policy, Social" OR "Social Policies" OR "Government Programs" [Mesh Terms] OR "Government Program" OR "Program, Government" OR "Programs, Government" OR "Government Sponsored Programs" OR "Government Sponsored Program" OR "Government-Sponsored Programs" OR "Government-Sponsored Program" OR "Social Welfare" [Mesh Terms] OR "Welfare, Social" OR "Services, Community" OR "Community Service" OR "Service, Community" OR "Community Services" OR "cash transfer" OR "conditional cash transfer" OR "cash transfer program" OR "in-kind" OR "financial incentive" OR "safety nets" OR "economic policy" #3 "Nutritional Status" [Mesh Terms] OR "Status, Nutritional" OR "Nutrition Status" OR "Status, Nutrition" OR "Anthropometry" [Mesh Terms] OR "Body Weights and Measures" [Mesh Terms] OR "Body Measures" OR "Body Measure" OR "Measure, Body" OR "Measures, Body" OR "Nutrition Assessment" [Mesh Terms] OR "Nutrition Indexes" OR "Indexes, Nutrition" OR "Nutrition Indices" OR "Nutritional Index" OR "Index, Nutritional" OR "Indices, Nutritional" OR "Nutritional Indices" OR "Nutrition Index" OR "Index, Nutrition" OR "Indices, Nutrition" OR "Premature Birth" [Mesh Terms] OR "Birth, Premature" OR "Births, Premature" OR "Premature Births" OR "Preterm Birth" OR "Birth, Preterm" OR "Births, Preterm" OR "Preterm Births" OR "Infant, Premature" [Mesh Terms] OR "Infants, Premature" OR "Premature Infant" OR "Preterm Infants" OR "Infant, Preterm" OR "Infants, Preterm" OR "Preterm Infant" OR "Premature Infants" OR "Neonatal Prematurity" OR "Prematurity, Neonatal" OR "Birth Weight" [Mesh Terms] OR "Birth Weights" OR "Weight, Birth" OR "Weights, Birth" OR "Growth" [Mesh Terms] OR "newborn weight" ditionalities, duration of transfers) and variable outcomes (anthropometric indices or measures, birthweight or gestational age).Data from the selected articles were extracted and entered into MS Excel.In cases of incomplete or missing data, the authors of the selected studies were contacted by email and asked to provide the requested information, or offer other clarification regarding the metrics evaluated.

evaluation of methodological quality
The methodological quality of the selected studies was assessed using the Quality Assessment Tool for the Dictionary of Quantitative Studies 25 , which classifies study quality according to risk of bias ("Strong", "Moderate" or "Weak").Two independent research authors assessed the risk of bias in each selected study.Disagreement was resolved by consensus or by consulting a third researcher (Table 1).Anthropometric outcomes: a difference of β= -0.14 [95%CI -0.27, -0.02] was identified in the H/A index among children aged up to 2 years with income < 1,000 reais, and β= -0.20 [95%CI -0.33, -0.88] among those with income > 1,000 reais.Regarding W/A, differences were β= -0.04 [95%CI -0.17-0.08]for the group with < 1,000 reais in income and β= -0.18 [95%CI -0.30, -0.05] for those with > 1,000 reais.Chakrabarti  Families received a fixed payment of US$ 2.623,45 for each child born or adopted Birthweight and gestacional age: the women who received the benefit had a much lower chance of having low-weight children in the future, while their proclivity to have another child (or the time of birth) remained unchanged.The effect is generally fueled by disfavored families (poor women, unmarried women, and women with low education.The transfer of earnings resulted in a drop of 0.9% and 0.7% in the category of children born with less than 2.000 and 1.500 g in low-income households over the next five years, representing a reduction of 49% and 83%, respectively.The effect on birth weight is pushed by premature babies, but there is no effect on the fracture of premature children.

Patient and public involvement
No patient involved.

Results
Searching performed in the literature returned 5,933 published studies.Of these, after removing duplicates, 5,825 articles were selected for title and abstract examination.In all, 42 studies were selected for textual analysis, with seven deemed eligible for inclusion.Additional manual searching involving the reading of other review articles produced two additional studies, resulting in a total of nine included studies (Figure 1).The search was updated on November 8, 2022 in all databases.615 new titles and abstracts were cheked.Of these, 3 texts were fully evaluated and there 2 studies met the inclusion criteria.
4][35] , while all three evaluated programs implemented direct cash transfers 9,[33][34][35] , one also provided health-related services 9 involving awareness activities and home visits targeting child health and nutrition, as well as community surveillance of childhood illness and acute malnutrition of mother-child pairs during each newborn's first 1,000 days of life.
Sample sizes ranged from 188 28 to 11,558 29 children under five years in seven studies focused on child development.As for four studies 30,31,35,34 investigating gestational age and birth weight, one 30 had a sample size of 55,998, the study 31 had 5,246,874 births; the study 35 had 14,591 births; and the study 34 had 19,274 births (Chart 3).

Methodological quality assessment
Both high (n = 3; 27.2%) and low (n = 4; 36,3%) methodological quality were identified in the evaluated studies, which were also categorized according to the following criteria: selection bias, study design, confounders, blinding, data collection method, and withdrawals and drop-outs (Chart 4).

Main results
Of the eleven included articles, seven analyzed CTs in low-and middle-income countries and evaluated child growth via anthropometry 9,[26][27][28][29]32,33 . All o these studies reported H/A ratio; W/H were reported in three 26,28,33 , while BMI/ A 32 and W/A (Sudfeld C. et al. 2021) 28 were evaluated in one each (Figure 2 and 3).
Three studies [33][34][35] , both investigating UCTs in high-income countries, investigated the effects of social protection programs on birth weight; two of these also considered prematurity 33,35 .
Another study in Brazil, conducted by the Center for Data and Knowledge Integration for Health, assessed the benefit of intergenerational transmission of health and poverty, as well as the relationship between PBF received by the mother and health and the health of the newborn.The authors present their findings that children born in a household where the mother received BF were less likely to have low birth weight (OR 0.93, CI; 0.92-0.94),very low birth weight (0.87, CI; 0.84-0.89),as well as to be born after 37 weeks of gestation (OR 0.98, CI; 0.97-0.99)or 28 weeks of gestation (OR 0.93, CI; 0.88-0.97) 31.
A cohort study carried out in India over an intervention period of 60 months evaluated the odds of stunting, as assessed by H/A among poor children under five years of age.The authors reported a reduction [from OR 0.89 (95%CI 0.81-0.98) to OR 0.85 (95%CI 0.77-0.94)]following the implementation of the Mamata CCT program across all Indian States (except Uttar Pradesh, Bihar, Jharkhand and Uttarakhand), as well as in the neighboring states of Odisha (West Bengal, Chhattisgarh, Jharkhand and Andhra Pradesh).

Discussion
The present systematic review synthesized the available evidence investigating the effects of CTs on child health outcomes in high-, low-, and middle-income countries.Our findings indicate that cash transfers are associated with reduced prematurity, low birth weight and improved nutritional status, as assessed by anthropometric indicators (W/A, H/A, W/H and BMI/A).The influence of CTs on the outcomes studied was mainly derived from CCTs whose effects were more pronounced among low-and middle-income countries.
Despite differences in scope, other reviews have also described positive effects on child health and nutrition outcomes resulting from CTs.Indeed, a recent review carried out by Anne E. Fuller et al. 36 reported better child health outcomes in families with children in Canada that received a CT.
Four studies evaluating birth outcomes (preterm birth, birth weight and hospital admissions during the neonatal period) provide evidence of the protective effects of UCTs on these outcomes 35 ; however, none evaluated the impact of CT on child growth.
A recent meta-analysis estimating the effect of CTs on diverse nutritional outcomes, as well as the proximal determinants of those outcomes, including diet quality and infant morbidity, concluded that after systematizing the results of 74 articles CTs were positively associated with higher H/A z-scores (HAZ) (p < 0.03) and a 2.1% reduction in stunting (p < 0.01).However, a similar effect was not observed for W/H (WHZ) (p < 0.42) and low weight z-scores (p < 0.07) 37 .The authors argued that the relatively short durations of the studies analyzed may not have been sufficient to evidence relevant changes in growth.When analyzed by region, significant impacts of CTs on W/A z-scores (WAZ) were only identified in Sub-Saharan Africa 37 , which is similar to the results of another study by Sudfeld C et al. 28 Two systematic reviews 37,38 suggest that UCTs positively impact child weight.The present review found that most studies examining the influence of CTs on child health demonstrated positive associations with birth weight, including a study by Briaux J et al. 9 , an RCT that combined monthly cash transfers with community activities targeting 1,035 mother-infant pairs during the first 1,000 days of life.These authors observed that female beneficiaries were less likely to have children with low birth weight (DiD = -11.8;ROR = 0.29; 95%CI 0.10-0.82;p = 0.02).However, Hamada R and Rehkopf D.H. 34 also evaluated UCTs and found no significant associations between cash transfers and birth weight (β = 18.0; 95%CI -17,8,53,8; p > 0.05).Nonetheless, the UCTs evaluated did appear to be positively associated with reductions in childhood illness, improved child weight and food consumption 37 .
With regard to assessment of the impact of CCTs on child nutritional outcomes in Latin America, such as linear growth, delayed child growth, and improvements in child health and nutritional status 39 , these programs were found to alleviate both poverty and food insecurity, in addition to bolstering school attendance and enhancing access to health services for beneficiaries 33,40,41 .
Among the recipients of CCTs in African countries, discordant results were reported 27,28 , as no significant effects on H/A were observed; however, the data did evidence positive effects on short stature, W/A, W/H and low weight.
A study by Kandpal E et al. 33 found a significant reduction in severe short stature in 6-36-month-old beneficiaries of the Philippine CT entitled Pantawid Pamilyang.Moreover, a study by Chakrabarti, Pan and Singh 29 identified a lower chance of stunting among children under five following the implementation of the Mamata Scheme in India.However, it is important to note that the nutritional findings reported in metanalyses and systematic reviews are not considered decisive, as no definitive evidence has been presented to conclusively document these effects 42,43 .
The impacts of CCTs on improved child health have been attributed to interventions related to health, nutrition and education.This has contributed to the success of these programs over time by enhancing beneficiaries' knowledge of important childcare practices and reinforcing the idea that, in the context of greater vulnerability, the provision of social benefits aims to contribute to improvements in child nutritional indicators 12,28,41 .
We additionally highlight that, despite potential improvements in populational health through the targeting of poor and vulnerable groups, the observed impacts of CCTs on child growth are not conclusive.Thus, it will be necessary to conduct studies examining the mechanisms underlying CCTs, especially in West Africa, where definitive evidence of program impact is lacking; moreover, the success seen in Latin America may not be replicable due to specific differences in CCT characteristics that may influence the effects of the studied results 9,27,28,35 .
According to studies, it is not possible to separate the reasons for which the CCTS is associated with better infant health outcomes.Suggestions are made that it is likely that women from families who receive benefits have received more appropriate prenatal care, which is linked to better outcomes 31,35 .
Regarding programs without conditionalities, three of the included studies 9,34,35 found UCTs to be associated with reduced low birth weight.Importantly, only one RCT by Briaux et al. 9 demonstrated a protective effect on H/A.
The present findings globally reinforce the impact of CT programs and suggest that attempting to mitigate short stature by means of a single interventional approach may prove difficult in at-risk communities that face a variety of contextual factors 9 .The positive results observed in H/ A 9 in the Togo study may be explained by the fact that the CT evaluated stunting and other forms of malnutrition in the mother-child binomial during the first 1,000 days of life, which likely maximized impact 16,44 .
The present review evidenced that child beneficiaries of the Columbian CT Familias en Acción presented increased BMI and a reduced the chance of being underweight; however, no impacts were observed on H/A, short stature, overweight or obesity among children aged two to five years 32 .In families benefiting from the Bolsa Família program in Brazil, this CT was negatively associated with H/A and W/A during the 24-month period studied 26 .
The authors argue that their findings can be explained by low participation in Bolsa Família, and probable errors in measurement of family income.Moreover, they maintain that, despite the lack of a direct association, the prevalence of short stature has progressively decreased in Brazil, particularly among poor families 26 .Another explanation may be that the relatively short duration of some studies may not have been sufficient to detect changes in linear growth, thus making it difficult to interpret the obtained results 9,37,39 .
Factors related to poverty, such as economic crises, austerity policies, food and nutrition insecurity and cutbacks in social protection programs, directly affect the health of children under five and impact infant mortality rates.A study carried out in Brazil found that the municipal level coverage of Bolsa Família was associated with significantly decreased mortality due to malnutrition (RR = 0.35; 95%CI 0.24-0.50) 5,10.
A strength of the present study was the adoption of a broad search strategy entailing the identification of published studies, reports employing robust methodologies and the absence of any language restrictions.In contrast to the focus of previous systematic reviews, the present work aimed to review the available evidence on the impact of CTs on child health outcomes, including anthropometry and prematurity.
Concomitantly, data in recent studies points to increasing rates of infant mortality in high-income countries [45][46][47] .The authors further speculate that this unusual finding is likely to be generalizable to other high-income nations in Western Europe and the US where associations between income and infant mortality have been evidenced 46,47 .
A study performed in England reported increased infant mortality mainly among socioeconomically disadvantaged children 47 .Academics have postulated that these increases may be due to recent cuts in health services and reductions in social benefits available to families 48,49 .
The present systematic review suffers from some limitations.First, a high degree of heterogeneity in eligible populations was observed, mainly in relation to the age of the children studied.Second, the CTs evaluated are highly variable in terms of design and duration, time of implementation and target population, which explains the inconsistencies in the estimated results, thus preventing the performance of a meta-analysis.Lastly, we excluded any studies that did not present results separately from those evaluating the effects of other social programs offering income, food and/or nutritional supplementation.
The results of the present systematic review indicate that cash transfer programs exert a positive effect on child growth as assessed by anthropometry and birth outcomes, thus affirming the use of CTs as a valuable social policy instrument for the promotion of child health.However, due to the small number of included studies herein, the body of evidence on this topic should be considered limited.
Accordingly, further study is needed to obtain additional clarification/confirmation and to allow for comparisons that would enable meta-analysis among studies.It would be interesting to elucidate, for example, whether specific positive findings identified among populations were linked to the effect of direct cash transfers, or whether these improvements resulted from the use of health services and/or by offering food and nutritional supplementation provided by other social programs.
The development of research aimed at analyzing and enhancing our understanding of nutritional dilemmas and the role of social policy interventions is important to protecting maternal and child nutrition and enhancing quality of life for future generations.It is therefore essential for forthcoming investigations to not only fully characterize the populations studied, but also to account for socioeconomic and demographic differences, as well as consider social determinants of health.
We must assume that these factors can impart differences in the magnitude and severity of nutritional status during childhood, and that the findings reported in the studies included herein are relevant, thereby affirming the notion that carrying out impact assessments on the effects of CTs on nutritional outcomes among more vulnerable populations continues to be necessary, further reinforcing the need for additional evidence on the role of social protections to mitigate short-and long-term consequences of malnutrition.

Collaborations
CS Lisboa, NS Guimarães, AJF Ferreira, FJO Alves, AS Rocha, N Ortelan, CSS Teixeira, IR Falcão, NJ Silva, RC Ribeiro-Silva, DB Santos and ML Barreto conceptualized and designed the study, drafted the initial manuscript, carried out the analysis plan, and reviewed and revised the manuscript.CS Lisboa, NS Guimarães, AJF Ferreira, KBB Silva, AS Rocha, N Ortelan, CSS Teixeira, NJ Silva, RC Ribeiro-Silva, DB Santos and ML Barreto conceptualized and designed the study and critically reviewed the intellectual content of the manuscript.All authors approved the final submitted version of this manuscript and accept accountability for all aspects of the work.

Figure 1 .
Figure 1.Flowchart detailing study search results and articles selected for systematic review.Excluded reports: Reason 1: different study population (n = 5): studies involving children aged over five years; Reason 2: distinct outcome (n = 3): studies investigating other health and nutritional outcomes, e.g., breastfeeding, vaccination; Reason 3: other type of study (n = 4): Studies not experimental in design; Reason 4: theoretical article (n = 4): studies lacking epidemiological approach; Reason 5: no cash transfer (n = 8): studies in which food or dietary supplements were offered; Reason 6: review article (n = 9); Reason 7: income and food transfer (n = 1): studies on programs offering cash as well as supplemental food; Reason 8: emergency cash transfer (n = 4).Source: Page et al., 2021 21 .

Chart 4 .
Characteristics of cash transfer programs and child growth and birth outcomes of studies included in systematic review (2016-2022).

Chart 3 .
Characteristics of all studies included in the systematic review(2016-2022).Characteristics of all studies included in the systematic review (2016-2022).Characteristics of cash transfer programs and child growth and birth outcomes of studies included in systematic review (2016-2022).

Table 1 .
Quality assessment of studies included in the present systematic review, in accordance with the Quality Assessment Tool for Quantitative Studies (2021-2022).