The Indirect Impact of Smallholder Vegetable Production on Children's Nutrition Outcomes in Rural Vietnam

Childhood undernutrition, particularly stunting, wasting, and micronutrient deficiencies, remains a major health concern in rural Vietnam. While literature suggests leveraging agriculture to improve child nutrition via agricultural diversification, market engagement, and women's empowerment, very few studies have empirically explored how smallholder vegetable production can influence household nutrition. This paper examines the association of household-level vegetable diversity, market access, and market participation with nutrition outcome measures of children in smallholder households. We use a cross-sectional household dataset, collected in 2016 in northwest Vietnam, covering 234 children aged 6–60 months. We estimate and compare the results of regression models using three-stage least squares (3SLS), ordinary least squares (OLS), logistic regression, and seemingly unrelated regression (SUR), to explore variations in six nutrition outcome measures: height-for-age z-score (HAZ), weight-for-height z-score (WHZ), weight-for-age z-score (WAZ), stunting, wasting, and underweight. Our results suggest smallholder vegetable production has a significant indirect association with children's nutrition status via market participation. Market participation is an important factor in improving girls' HAZ and WHZ, and in reducing the probability of boys being stunted and underweight. The additional income from selling vegetables allows households to purchase nutritious food, which is likely to have a positive impact on children's nutrition outcomes.

Introduction

Vietnam has shown significant progress in reducing children's undernutrition in Southeast Asia [Association of Southeast Asian Nations (ASEAN) et al., 2016], but this has decelerated in recent years. The prevalence of stunting and underweight went down by 50 percent between 2001 and 2010 (Chaparro et al., 2014). However, between 2011 and 2015 there has been a relative increase in childhood stunting (from 23 to 25%) and wasting (from 4 to 6%) (Global Nutrition Report, 2022). Current interventions focus on the provision of key micronutrients, such as vitamin A, iron, and zinc (Chaparro et al., 2014), because children's undernutrition is due to micronutrient deficiencies arising from poor dietary intake and from infectious diseases like diarrhea [United Nations Children's Fund (UNICEF), 2013]. Current estimates indicate about 14, 29, and 52% of young children are vitamin A-, iron-, and zinc-deficient, respectively (Chaparro et al., 2014). However, declining financial support from the national and local governments for these activities, lack of strategic oversight and reduction of international aid, and the poor nutritional knowledge of mothers and other household members (Socialist Republic of Vietnam, 2012; IFPRI, 2016; Harris et al., 2021), have impeded progress. The positive impacts of these changes also differ between urban and rural areas, by region, and between the Kinh majority and ethnic minorities. Women, children, and ethnic minorities are the most vulnerable groups (National Institute of Nutrition and United Nations Children's Fund, 2011).

One way to lessen the financial burden of rural households is to promote agriculture and other existing production systems in locations with high populations of these vulnerable groups (Alderman et al., 2013). The agriculture sector therefore can play a crucial in combatting child undernutrition by addressing inadequate access to nutrient-rich food (Sharma et al., 2021). In Vietnam, many undernourished children are found in resource-poor households in rural areas dependent on agriculture for their main source of livelihood and nutrition (Linh and Glewwe, 2011).

Recent agriculture-nutrition projects have integrated a generalized theory of change, e.g., increasing production, income, women's empowerment, and/or nutrition knowledge to improve food security and dietary consumption, leading to improvements in nutritional status of target beneficiaries (Herforth and Ballard, 2016). At the household level, the pathway between agriculture and children's nutrition have been linked via the direct and indirect pathways: (1) household production of nutrient-rich agricultural products which may subsequently be consumed by the household members; and (2) market engagement where at least a share of the nutrient-rich agricultural products produced by the household are sold at markets generating agricultural income used to purchase diverse food products, or access to a wide variety of food due to market proximity (Dangour et al., 2012; Gillespie et al., 2012; Turner et al., 2013; Herforth and Harris, 2014; Kadiyala et al., 2014; Pinstrup-Andersen, 2014; Kanter et al., 2015; Sharma et al., 2021). In the direct pathway (pathway 1), the diversity and specific mix of agricultural crops cultivated by a household can influence household members' dietary quality and eventually nutrition. This is especially true for subsistence households that only consume what they produce (Muller, 2009). The indirect pathway (pathway 2) relates to how poor market access and low household incomes can be interrelated, and its effect on dietary diversity and nutrition. However, when households have increased market access and opportunities for market participation, they are faced with complex decisions about which agricultural crops to produce, consume and sell, and how to use any additional income generated from the sale of agricultural crops. Nevertheless, recent studies have highlighted the importance of market engagement because its effect on household nutrition has been found to be larger than agricultural diversity per se (e.g., Hirvonen and Hoddinott, 2017; Koppmair et al., 2017; Stifel and Minten, 2017), though much remains unknown about the impacts of market engagement and household vegetable production more broadly.

In addition to these pathways, other factors can also affect dietary diversity and, ultimately, child nutrition [Black et al., 2013; United Nations Children's Fund (UNICEF), 2013]. For instance, the role women play in resource-poor households affects the diets of household members (Alderman et al., 2013). As women tend to be more concerned with the health and well-being of children and other household members, their ability to control resources at home, to have access to credit, and to spend time selecting and preparing diverse and nutritious food for the family can have a strong, positive effect on household dietary quality, especially on children's diets (Quisumbing et al., 1995). Likewise, intrahousehold food allocation may be unequally distributed due to age and gender biases, and seasonal food shortages (Abdullah and Wheeler, 1985; Gittelsohn, 1991; Luo et al., 2001). Girls and women of reproductive age in certain countries in East and South Asia are often discriminated against when it comes to intrahousehold food allocation (Gittelsohn, 1991; Luo et al., 2001; Murphy et al., 2011; Madjdian and Bras, 2016; Rahman, 2018). In some rural villages in Bangladesh and the Philippines, children under 5 years old and/or women receive proportionately lower energy and protein intakes compared to the household's adult males (Abdullah and Wheeler, 1985; Senauer et al., 1988). In Indonesia, mothers consumed more carbohydrates than other family members (Wibowo et al., 2015). Girls and women of reproductive age in certain countries in East and South Asia are often discriminated against when it comes to intrahousehold food allocation (Gittelsohn, 1991; Luo et al., 2001; Murphy et al., 2011; Madjdian and Bras, 2016; Rahman, 2018). These associations are further influenced by other confounding factors like geography, ethnicity, education, maternal care practices, access to government supplementary feeding and immunization programs, WASH (water, sanitation and hygiene) practices, among others.

Vegetable consumption is a particularly sustainable way to combat some types of child undernutrition because of its micronutrient richness. However, its overall impact on child undernutrition remains largely unexplored. Most studies that have evaluated the impact of vegetables on nutritional outcomes were part of horticultural intervention programs (see review by Taren and Alaofe, 2013), which produced mixed findings (Masset et al., 2012; Ruel et al., 2013; Carletto et al., 2015).

This study therefore aims to fill this gap by examining the possible contribution of smallholder vegetable production on children's nutrition and identifying other significant factors. We investigate these relationships using pooled and gender-disaggregated samples. We incorporate four indicators of smallholder vegetable production in the standard empirical approach of examining child nutrition, controlling for the underlying child, maternal (including women's empowerment), and household characteristics.

We use cross-sectional data to study 234 children, aged 6–60 months, from 188 rural households in northwest Vietnam. We hypothesize that smallholder vegetable production has a positive effect on children's nutrition, which may occur either through a direct or indirect pathway. In the direct pathway, we specifically investigate whether diverse vegetable production is associated with childhood nutrition. The premise is that when households diversify their production, children also diversity their food intake accordingly, and therefore improve their diet, ultimately leading to improved nutrition. In the indirect pathway, we assess whether market engagement, via market access and market participation, positively correlate with nutritional outcomes. Households close to markets enjoy a wider array of diverse and nutritious foods that are not accessible to those in more remote rural areas. Similarly, the consumption of more diverse foods, leading to improved nutritional status, becomes possible because of income gains due to semi-commercialization (market participation). Understanding these direct and indirect pathways can help the Government develop long-term, sustainable solutions that are resource-efficient, and allow households to more effectively participate in.

Materials and Methods

Study Area

The study was conducted in Lao Cai Province, a temperate vegetable-producing area in Vietnam's Northern Midlands and Mountainous region. It is an ideal research location for several reasons. Firstly, the province is among the poorest in Vietnam, with 22 percent of its population living below the poverty line (Lao Cai People's Committee, 2016). Secondly, the region is home to many ethnic minority communities, which are considered to be among the most vulnerable groups in the country (Kozel, 2014). They comprise 66 percent of the province's population: Mong (25%), Tay (15%), Dao (14%), Giay (5%), and Nung (4%) [General Statistics Office (GSO) of Vietnam, 2020]. Generally, they show preference for sons, due to their patriarchal and patrilocal clan system (Jones et al., 2014). Boys play significant cultural and economic roles, especially in religious rituals, in continuing the family heritage and in providing support to parents during old age (Jones et al., 2014).

Farming constitutes 93 percent of the economy in this region, especially in its rural areas (Linh and Glewwe, 2011). Many rural households grow vegetables only for their own consumption due to small landholdings [International Food Policy Research Institute (IFPRI), 2002]. Nevertheless, the province is suited to increasing vegetable intensification due to its climate and soil suitability (Wijk and Everaarts, 2007). In 2019, the province's total vegetable output was 11.74 tons per hectare [General Statistics Office (GSO) of Vietnam, 2020] compared to the country average of 16.82 tons per hectare [Food and Agriculture Organization Corporate Statistical Database (FAOSTAT), 2022]. In spite of this, fruit and vegetable consumption comprise only 70 grams/capita/day [International Livestock Research Institute (ILRI), 2014], which is below the daily recommendation of 400 grams/capita/day (World Health Organization and Food and Agriculture Organization of the United Nations, 2005). In the province, diets are mainly composed of four food groups (cereals, vegetables, meat and poultry, and nuts and beans), which is lower than the national average of 10–11 food groups (Kim et al., 2021; Genova et al., 2022). Government nutrition programs have already been implemented in the province to overcome nutritional issues (Chaparro et al., 2014). However, among children under five, it still has one of the highest rates of stunting (31%) and underweight (18%) due to inadequate household food security (caused by poverty) and inadequate maternal and childcare (National Institute of Nutrition and United Nations Children's Fund, 2011; National Institute of Nutrition, 2019).

Fourthly, the region has the highest rates of child marriage (19%), where women marry before their eighteenth birthday [United Nations Population Fund (UNFPA) and United Nations Children's Fund (UNICEF), 2018]. Evidence shows that early marriages and premature childbearing can have a negative effect on maternal health (Goli et al., 2015), which can be transferred to her offspring (Parsons et al., 2015).

Lastly, many households are poor due to a “spatial trap”; i.e., their remoteness limits their access to resources like health, education, infrastructure and credit, and they may be isolated from both input and output markets (Epprecht et al., 2009; Dang, 2012; Kozel, 2014). Much of the land is considered sloping (Figure 1). This geographic diversity has resulted to ethnic minorities developing their own distinctive traditional farming systems, cultural practices, and indigenous knowledge that are well-suited to the environment in which they live in Tran (2003). This has also led to low agricultural productivity. Their strong adherence to cultural traditions versus acceptance of new agronomic technological innovations that offer opportunities to improve productivity and efficiency is a major factor to their poverty (Tinh, 2002; Bonnin and Turner, 2012).

FIGURE 1

Figure 1. Administrative map of Lao Cai province. Source: SRTM downloaded from http://srtm.csi.cgiar.org/SELECTION/inputCoord.asp; Data: SRTM 90m digital Elevation Model (DEM) using ArcGIS10 http://srtm.csi.cgiar.org/SELECTION/listImages.asp.

Study Design

The authors gathered the data set analyzed in this study as part of an extensive rural household survey conducted in July-August 2016 and November 2016 in Lao Cai province, Vietnam. This study was part of a large project “Toward more profitable and sustainable vegetable farming systems in north-western Vietnam (AGB/2012/059),” funded by the Australian Centre for International Agricultural Research (ACIAR), aiming to improve the livelihoods of smallholder farmers in north-western Vietnam through market engagement and improved resource and disease-management practices. This study contributes to the broader aim of the ACIAR project by investigating whether there is a relationship between smallholder household characteristics and diets and nutritional outcomes. Ultimately, the goal of the research is to contribute information that may lead to programs and policies that address undernutrition among children in Lao Cai province, and improve livelihoods of the population in these remote areas in the long run.

Data was collected in two time periods to capture possible seasonal variations in both household consumption and household vegetable production. The survey sample included households from four districts (Bac Ha, Sa Pa, Muong Khuong, and Si Ma Cai). These districts were selected based on each having a mean elevation of 600 meters or more above sea level (masl) as temperate vegetable cultivation is best achieved at higher altitudes. Figure 1 shows the administrative map of the province that is color-coded to indicate geographic heterogeneity.

Smallholder households were selected using a stratified multistage sampling strategy. Households had to be engaged in agricultural production (of any product, not just vegetables) during the past 3 years to qualify for the study. Communes in the four selected districts were grouped using the median elevation and median vegetable density per capita data. From the elevation and vegetable density per capita data, each commune was coded as either low elevation (≤1,335 m) or high elevation (>1,335 m), and low vegetable density per capita (≤98) or high vegetable density per capita (>98). Then, communes were categorized into four strata: low vegetable density per capita-low elevation (LV-LA); low vegetable density per capita-high elevation (LV-HA); high vegetable density per capita-low evaluation (HV-LA); and high vegetable density per capita-high elevation (HV-HA). Finally, three communes in each stratum and four villages per commune were randomly selected¹. In each village, ten households were randomly selected from a list of households provided by the rural commune (xã)/commune-level town (thị trấn) administrative officers, yielding 510 households from 51 villages in 13 communes.

Data

Our analyses focused on households with children aged 6–60 months. Detailed information on household and individual characteristics—including anthropometric measurements, intrahousehold food consumption, food choice and dietary beliefs, women's empowerment (using the modified Abbreviated-Women's Empowerment in Agriculture Index or A-WEAI Malapit et al., 2017), micronutrient supplementation and maternal nutritional knowledge, WASH practices, household income and expenditure, farm production and marketing activities in the last production cycle, and child health— were collected. Except for the women's empowerment module, the person primarily responsible for food preparation in the household was the target respondent, which in most cases is the child's mother. The modified A-WEAI module was administered to both the principal and secondary decision-makers in the household (normally, the parents of the child).

The semi-structured questionnaire was translated into Vietnamese by two native Vietnamese speakers and was validated by the Mekong Development Research Institute (MDRI)². Pre-testing was done in the study areas in March, June, and July 2016. The paper-based questionnaire was transferred to a mobile application (CommCare³). A detailed Training Manual was provided to a team of 22 enumerators and five field supervisors from MDRI who facilitated the survey implementation. The team included one local enumerator fluent in the local Mong dialect. Enumerator training was conducted over a week and modifications were made to the survey instrument.

Anthropometric Measurements

Children were measured using the anthropometric protocol from the National Health and Nutrition Examination Survey (NHANES) Anthropometry Procedures Manual [Centers for Disease Control and Prevention (CDC), 2016]. The seca^® 210 infantometer was used to record the recumbent length of infants and children up to 3 years who cannot stand on their own. The height of older children was measured using a tape measure⁴. The Laica® digital weight scale was used to record children's weight. The weight of infants and young children unable to stand independently was recorded with the aid of the mother. The mother and the child were weighed together, and then the mother's weight was deducted to derive the child's weight. Mothers were requested to remove their and their children's, shoes, hats, and heavy clothing, prior to the height and weight recording.

Using the 2006 WHO Child Growth Standards and the Stata user-written program zscore06 (Leroy, 2011), children's height and weight were converted into z-scores: height-for-age z-score (HAZ), weight-for-height z-score (WHZ), and weight-for-age z-score (WAZ). Three additional variables were created to measure the prevalence of stunting, wasting, and underweight, using the HAZ, WHZ, and WAZ, respectively. WAZ is a composite measure of HAZ and WHZ, and is difficult to interpret. However, it is commonly used to monitor growth and changes in malnutrition over time (O'donnell et al., 2008). Underweight is therefore used as a summary indicator, reflecting a child being stunted or wasted (Kimenju and Qaim, 2016). Children with missing data on height, weight, age, and gender were excluded from our analyses. The observed standard deviations of the z-scores can also identify inaccurate data due to measurement or coding errors. Therefore, we followed the WHO cut-off points for each of the child nutrition outcomes to identify these errors and exclude children whose values are in these ranges: HAZ, ≤-6 or ≥6 SD; WHZ, ≤-5 SD or ≥5 SD; and WAZ, ≤-6 or ≥5 SD [World Health Organization (WHO), 2010b].

Econometric Model Specification

We adopt the conceptual framework commonly used to explain the causes of child undernutrition [Black et al., 2013; United Nations Children's Fund (UNICEF), 2013]. In this framework, poor diets and diseases are the immediate determinants of child undernutrition. These factors, which often occur together, are caused by several underlying factors, like lack of access to food, either physically or economically, poor maternal and child-care practices, and an unhealthy home environment, due largely to income poverty.

The empirical model estimated is shown in equation (1):

$\begin{array}{l}{H}_{ijh}=constant+\beta VegDiversit{y}_h+\gamma Marke{t}_h+{\delta }^{\prime }{\text{G}}_h\\ +{\omega }^{\prime }{\text{C}\text{H}}_i+{\sigma }^{\prime }{\text{M}}_j+{\phi }^{\prime }{\text{H}\text{H}}_h+{\epsilon }_{ijh}& (1)\end{array}$

where H_ijh is one of the six alternative nutrition outcome measures: HAZ, WHZ, WAZ, stunting, wasting, and underweight of child i of mother j in household h. The HAZ, WHZ, and WAZ are continuous variables, and the prevalence of stunting, wasting, and underweight are binary variables. G_h, CH_i, M_j, and HH_h are vectors of women's empowerment/adequacy, child characteristics, maternal characteristics, and household features, respectively; and ε is the error term. In these models, we focus on the estimates of β and γ, which indicate whether our main variables of interest have positive, negative, or no effects on the specific nutrition outcome measure. The main variables of interest and specific controls used in the model are explained subsequently.

We hypothesize that smallholder vegetable production may have a positive relationship with children's nutrition. The link between household vegetable production and children's nutrition could be a result of either consumption from household production (direct), or consumption of more diverse foods due to market access and/or higher income gained from market participation (indirect) (e.g., Carletto et al., 2015; Kanter et al., 2015). We measure the association of smallholder vegetable production and children's nutrition outcomes using four main variables of interest: VegDiversity, TimeMarket, TradMarket, and ModMarket.

VegDiversity_h is a measure of household vegetable production diversity, and is a simple unweighted count of the number of unique vegetables cultivated by household h in the last 12 months prior to July/August 2016 (Keding et al., 2012). VegDiversity_h was assessed by asking the respondent about the types of vegetable cultivated for home consumption, for market, lost due to postharvest spoilage, given as gift to neighbors or used for seed storage. We hypothesize that greater diversity in vegetable production can improve dietary quality and, eventually, nutrition through the subsistence pathway (Sibhatu and Qaim, 2017).

Recent studies have also pointed to the importance of markets (Market_h) on children's nutritional status, in terms of both access and market participation (e.g., Hirvonen and Hoddinott, 2017; Stifel and Minten, 2017). In this case, we include TimeMarket to indicate market access, and TradMarket and ModMarket to indicate market participation. TimeMarket is the one-way travel time in hours to the nearest food market. We have added this variable to differentiate remote rural households from those in closer proximity to food markets. We hypothesize that closer proximity to food markets allows children to have more diversified food intake, which may explain improvements in the nutritional status of children living closer to markets as compared to children who are residing in remote areas (Abay and Hirvonen, 2016). We impute the missing values and “don't know” responses with the mean of the village, commune, or the district, depending on which is available, by transport type. Two binary variables indicate market participation in the last production cycle: if they have sold to traditional channels (TradMarket), and to modern channels (ModMarket). Traditional channels are the local market in the village/commune/district, selling to fellow farmers, and selling to collectors that visit the farm(s). Modern market channels include supermarkets, wholesale markets, cooperatives, and other retailers in Hanoi and other nearby provinces. For these two binary variables, our hypothesis is that market participation positively affects children's nutritional status through the income gained from semi-commercialization, allowing the household to purchase more diversified food, thus promoting better nutritional status.

We also control for women's empowerment/adequacy (G_h) since previous studies suggest that women's empowerment has been a significant predictor of children's nutritional status. Women are normally more health-promoting than men through the preparation of more nutritious and healthier food, and through childcare (Quisumbing et al., 1995). For instance, in Burkina Faso, improvements in women's empowerment (spousal communication, purchasing decisions, healthcare decisions, and family planning decisions) contributed to the program's impact on reducing wasting (Heckert et al., 2019). A recent study also found positive association between women's empowerment and gender equality with child HAZ using the Feed the Future population-based survey data from Bangladesh, Cambodia, Ghana, Mozambique, and Tanzania (Quisumbing et al., 2021). In this study, two binary variables, which are adopted from the A-WEAI indicators, are used to control for women's empowerment/adequacy: woman make decisions on household income (Income), and woman's workload is less than the time poverty line of 10.50 h (Workload). Each indicator is assigned a value of 1 if the woman's achievement is adequate, i.e., it exceeds the defined inadequacy cut-off for the specific indicator, and a value of 0 if otherwise (Malapit et al., 2017). Workload indicates reduced workload, which could indicate better maternal care through mothers having sufficient time to make healthier decisions about their children's food intake, grooming, and sanitation (Cunningham et al., 2019).

CH_i are child-related covariates that include dietary diversity (DDS), age (AgeChild); and binary variables for male child (MaleChild) and incidence of diarrhea in the 2 weeks prior to the interview (Diarrhea). DDS is defined as a simple unweighted count of the number of unique food groups consumed by the child in each reference period. In this study, the DDS ranges from 1 to 14, and is based on the 14 food groups in the Vietnamese Food Composition Table 2007 edition⁵: cereal and products; starchy roots and tubers; nuts and beans; vegetables; fruits and berries; oils, fats and butters; meat, poultry game and its products; fish, shellfish, and products; eggs and products; milk and milk products; canned products; sugars, preserves and confectionery; soft drinks and other beverages including alcoholic drinks; and spices and sauces. The derivation of the DDS considers all food items consumed by the child in each reference period. These food items are classified three ways: (a) home-cooked meals; and food sourced elsewhere, either purchased or provided by relatives and/or neighbors, composed of (b) cooked meals, and (c) specific food items (uncooked) that were directly consumed by children, e.g., bread, branded milk, fresh fruits and raw vegetables, candies, chips, and juices. DDS is the mean Dietary Diversity Score of the child in three reference periods. We include Diarrhea since it has been shown to negatively impact on children's nutritional outcomes, especially WHZ and wasting [World Health Organization (WHO), 2010a].

M_j are maternal covariates that include age (AgeMother), education (EducMother), height (HtMother), body mass index (BMIMother), and a binary for Mong ethnicity (MongMother). EducMother was highly correlated with the nutritional knowledge of the mother so we treat it as proxy for better child-care practices and greater maternal nutritional knowledge (Alderman and Headey, 2017). A mother's height indicates her genetic endowment; women who are stunted are likely to have stunted offspring (Addo et al., 2013; Forero-Ramirez et al., 2014; Prendergast and Humphrey, 2014; De Onis and Branca, 2016; Wu et al., 2021). We also included BMI because it has been shown to effect the likelihood of child stunting (Headey et al., 2012).

HH_h are household-level covariates that include geographic stratum to which the village belongs to according to elevation and vegetable per capita density⁶ (LV-LA, LV-HA, HV-LA, and HV-HA), number of children <5 years old (CHD5b), total cultivated area (Area), monthly non-food expenditure per capita (NonFoodExp_pc), and a binary variable for access to improved toilet facility (ImprovedToilet). The geographic stratification was included to control for geographical differences affecting vegetable production due to elevation or other factors. CHD5b is a proxy for the mother's ability to undertake good child-care practices. The more infants and younger children in the household, the more time constrained the mother is as she juggles childcare and other household (including agricultural) activities (Babu et al., 2014). Area is a control for farm-specific characteristic, while NonFoodExp_pc and ImprovedToilet⁷ are proxies for household income and good sanitation, respectively.

Empirical Strategy

We first estimate equation (1) using ordinary least-squares (OLS) regressions for the three continuous nutrition outcome variables (HAZ, WHZ, and WAZ) and Logit regressions for the three binary outcome variables (stunting, wasting, and underweight) to assess the empirical relationship between children's nutritional outcomes and smallholder vegetable production. We present the marginal effects for the Logit estimations. In both estimators, robust standard errors are used and are clustered at the household level.

Three of our main variables of interest— VegDiversity, TradMarket and ModMarket— are potentially endogenous. This renders the OLS and Logit biased and inconsistent estimators. Using Wooldridge's robust score chi² and robust regression-based F-test for exogeneity⁸, we find that, in some of the models, endogeneity exists with TradMarket and ModMarket, and hence the instrumental variable (IV) method is preferred. Therefore, we use the percentage of surveyed neighbors in the village selling to traditional channels, excluding the household (PctNeighborTrad) as an instrument for TradMarket; and the percentage of surveyed neighbors in the village selling to modern channels, excluding the household (PctNeighborMod) for ModMarket. The use of the percentage of neighbors selling to traditional channels excluding the household (PctNeighborTrad) is a good instrument for TradMarket since a household selling to TradMarket is likely to have neighbors similarly engaged in selling to TradMarket (Andersson et al., 2015). The same can be said for PctNeighborMod.

Mathematically, we model them as follows:

$\begin{array}{l}VegDiversit{y}_h^V={\beta }_0^V+{\beta }_1^{V}Distance+{\beta }_2^V{G}_h+{\beta }_3^{V}C{H}_i\\ +{\beta }_4^V{M}_j+{\beta }_5^{V}H{H}_h+{\epsilon }_h^V& (2)\end{array}$

$\begin{array}{l}TradMarke{t}_h^T={\beta }_0^T+{\beta }_1^{T}PctNeighborTrad+{\beta }_2^T{G}_h\\ +{\beta }_3^{T}C{H}_i+{\beta }_4^T{M}_j+{\beta }_5^{T}H{H}_h+{\epsilon }_h^T& (3)\end{array}$

$\begin{array}{l}ModMarke{t}_h^M={\beta }_0^M+{\beta }_1^{M}PctNeighborMod+{\beta }_2^M{G}_h\\ +{\beta }_3^{M}C{H}_i+{\beta }_4^M{M}_j+{\beta }_5^{M}H{H}_h+{\epsilon }_h^M& (4)\end{array}$

Equations (2)–(4) form a system of equations that account for the endogeneity of VegDiversity⁹, TradMarket and ModMarket. We use the three-stage least squares (3SLS) method in all six alternative nutritional outcome measures, which is robust to the application of instrumental variables (IV) in the seemingly unrelated regression (SUR), since error terms of the system of equations might be correlated. The application of linear regression when the dependent variable is dichotomous, in this case the three prevalence outcome variables (stunting, wasting, underweight), has been justified by Angrist and Pischke (2009) and Hellevik (2009). Since the error terms might be correlated between the system of equations for a specific child and be uncorrelated across children, the seemingly unrelated regression (SUR) models are also estimated. The Breusch-Pagan test of independence indicated statistically significant correlations between the error terms of the system of equations in all six alternative nutritional outcomes. Therefore, the main results of the 3SLS estimates are used for inference in the subsequent section, and are compared with OLS, Logit, and SUR estimates to determine the possible direction of the bias.

We use Stata v15.1 (StataCorp, College Station, TX) for the descriptive and statistical analyses. Analyses are done for the pooled sample and were gender-disaggregated to identify any differential effects for boys and girls.

Results and Discussion

Descriptive Statistics

Our sample consists of 266 children aged 6–60 months from 210 households. After cleaning the data, the final number of observations with complete information is 234 children from 188 households. We have a nearly balanced sample of boys (n = 123) and girls (n = 111). Table 1 provides the definition and summary statistics of all variables used in the analyses.

TABLE 1

Table 1. Descriptive statistics of key variables of children 6–60 months in Lao Cai province, Vietnam.

The main variables of interest comprise variables that measure vegetable production diversity (VegDiversity), market access (TimeMarket), and market participation (TradMarket and ModMarket). The mean VegDiversity is 3, and ranges between 0 and 13 vegetables. TimeMarket is 0.39 h on average, with the most remote household traveling for 1.50 h to access the market. Most of the households are involved in vegetable production, with 60 percent having sold to some type of market (44% to TradMarket, 20% to ModMarket, and about 5% to both traditional and modern channels).

More than 60 percent of woman decision-makers make decisions on Income and nearly 30 percent have a workload that is less than the time poverty line of 10.50 h per day (Workload).

The average age of young children in our sample is 37.40 months, and they consume food belonging to four food groups (DDS), on average. Seventeen percent of the 188 households indicate that some or all of their children have experienced diarrhea 2 weeks prior to the survey (Diarrhea).

Most mothers belong to the Mong ethnic group, and their average age is 41 years. The oldest is 87 years old, because the respondent is the child's grandmother; the biological mother is working elsewhere or has passed away, and the grandmother took over the mother's role. Mothers who are working and residing elsewhere are excluded since we only include household members staying in the house 4 days of the week or 6 months of the year. About 40 percent are illiterate, 22 percent have primary-level education, and 33 percent have reached secondary-level¹⁰. The average maternal height is 149.77 centimeters (cm), which is considered stunted following the definition of Addo et al. (2013) on maternal stunting as being below 150.10 cm. The mean BMI is 22.40 kilograms per square meter (kg/m²), which is considered “normal” weight (WHO Expert Consultation, 2004).

The average number of children under 5 years old is two. We find high variability in the total cultivated land area (Area); the average is 1.08 hectares, and the largest land area is 19.60 hectares. Wide variation is also observed for NonFoodExp_pc, with the lowest monthly non-food expenditure per capita equal to 0.04 million VND¹¹ (US$1.72) and the highest is 13.88 million VND (US$598.06). The majority of households do not have access to a safe water source and drinking water, and only about 37 percent have access to improved toilet facilities.

The mean HAZ, WHZ, and WAZ are −2.20, −0.25, and −1.44 SD below the medians in the reference population, respectively (Table 1). Figures 2, 3 show the overall distribution for HAZ, WHZ, and WAZ, for the pooled sample and by gender, respectively. These figures further show the extent of stunting, where the majority of children have HAZ below the WHO child growth standards depicted as standard normal densities. In the pooled sample, 62 percent are stunted (HAZ < -2 SD), 9 percent are wasted (WHZ < -2 SD), and 24 percent are underweight (WAZ < -2 SD) (Figure 3). These estimates are higher than the 2019 provincial and regional averages (National Institute of Nutrition, 2019) and are consistent in each of the four districts. By gender, there are slightly more boys who are stunted (65%) and underweight (24%), and more girls who suffer from wasting (13%) (Figure 3). The higher prevalence of stunting among boys was also observed in several countries in Africa, which suggests that boys are more vulnerable to health inequalities than girls (Wamani et al., 2007; Thurstans et al., 2020).

FIGURE 2

Figure 2. HAZ, WHZ, and WAZ of sampled children in Lao Cai province. Note: WHO child growth standards are shown in standard normal densities. Source: Vietnam Rural Household Survey (July–August 2016).

FIGURE 3

Figure 3. HAZ, WHZ, and WAZ of sampled boys (A) and girls (B) in Lao Cai province. Note: WHO child growth standards are shown in standard normal densities. Source: Vietnam Rural Household Survey (July–August 2016).

Disaggregating them by age group, we find a higher incidence of stunting in the 24–60-month group, and a higher incidence of wasting in the 6–23-month group. The results for the children in the 6–23-month group may represent growth faltering, while those for the 24–60-month group may reflect the full impact of various postnatal issues (Alderman and Headey, 2018). We find more girls who are stunted (64%) and more boys who are wasted and underweight (20 and 28%, respectively) in the 6–23-month old age group. In the 24–60-month old age group, stunting (67%) and underweight (28%) are more prevalent among boys, and wasting is more prevalent among girls (27%). By geographic strata, stunting and underweight are highest in HV-LA (70 and 30%, respectively). In contrast, the lowest incidence of stunting and underweight are in HV-HA (49%) and in LV-LA (18%), respectively. In the subsequent analyses, results are presented for the pooled sample and disaggregated by gender, instead of by age group, due to small sample size.

Impact of Smallholder Vegetable Production on HAZ, WHZ, and WAZ

In the pooled sample, only market participation (TradMarket and ModMarket) is significantly associated with improvements in WAZ (Table 2, Panel 9). Households that sell to TradMarket benefit via a WAZ increase of 0.334 SD (p < 0.10). The association is larger and more significant when households sell to ModMarket, with a WAZ increase of 0.354 SD (p < 0.05). These associations are robust across different estimation methods (OLS and SUR).

TABLE 2

Table 2. Impact of smallholder vegetable production on HAZ, WHZ, and WAZ of pooled sample.

By gender, VegDiversity and market participation (TradMarket and ModMarket) are significantly associated with HAZ, WHZ, and WAZ, especially for girls. VegDiversity is associated with a 1.134 SD increase in HAZ (p < 0.05) among girls (Table 3, Panel 7). However, it affects the WHZ (Panel 8) of girls and boys differently. For girls, a 1.201 SD decrease in WHZ (p < 0.05) is observed, while a 0.772 SD marginal increase (p < 0.10) is observed for boys. For market participation (TradMarket and ModMarket), significant associations were only found for girls, with larger coefficients and more significant positive effects of market participation on HAZ and WHZ (p < 0.05), and WAZ at the 5- and 10-percent levels of significance (Panels 7 to 9). The positive and significant association between ModMarket and WAZ is robust across different estimation techniques, similar with the findings in the pooled sample (Table 2, Panel 9). In particular, ModMarket is associated with a 1.115 SD increase in HAZ (p < 0.05), and a 0.487 SD increase in WAZ (p < 0.10). TradMarket is also associated with a 0.905 SD increase in WHZ (p < 0.05), and a 0.498 SD increase in WAZ (p < 0.05).

TABLE 3

Table 3. Impact of smallholder vegetable production on HAZ, WHZ, and WAZ of sampled boys and girls.

These results suggest that market participation does have a significant effect in improving the HAZ, WHZ, and WAZ of children in this province, especially among girls. Selling to traditional channels can improve the current WHZ. The effect is greater when households engage in modern market participation; it can positively affects children's linear growth (HAZ), an indicator of long-term undernutrition. Also, the link between vegetable production diversity and children's nutritional status should not be discounted. While its impact differs in HAZ and WHZ, what is crucial is that, like ModMarket, it is also associated with significant improvements in young girl's linear growth (HAZ). This has important implication for curbing the transmission of intergenerational linear growth retardation (Walker et al., 2015) in rural areas, as we find that a one unit increase in maternal height decreases the probability of stunting (Table 4, Panel 7, p < 0.05).

TABLE 4

Table 4. Impact of smallholder vegetable production on the prevalence of stunting, wasting, and underweight of pooled sample.

Impact of Smallholder Vegetable Production on the Prevalence of Stunting, Wasting, and Underweight

In general, TradMarket and ModMarket have implications for reducing the probability of underweight and stunting, respectively (Table 4, Panels 7 and 9). Increased VegDiversity, however, is associated with an increased probability of being underweight although the association is marginal (Panel 9).

By gender, VegDiversity and market participation (TradMarket and ModMarket) are significantly associated with stunting, wasting, and underweight, especially for boys. VegDiversity, while reducing the probability of wasting (Table 5, Panel 8), does not reduce the probability of stunting and underweight (Panels 7 and 9, respectively) among boys as initially hypothesized. TradMarket and ModMarket are strongly associated with a reduced probability of a boy being stunted (Panel 7, p < 0.05). However, ModMarket also leads to a higher prevalence of wasting (p < 0.05) but this association requires further validation because the number of observations of wasted boys is only six percent.

TABLE 5

Table 5. Impact of smallholder vegetable production on the prevalence of stunting, wasting, and underweight of sampled boys and girls.

This strengthens the finding that additional income gained from market participation may improve child nutrition outcomes. It also highlights that by simply selling to traditional channels rather than not selling, there is a strong and positive link with a reduced probability of being underweight.

Other Factors Associated With HAZ/Stunting and WHZ/Wasting Outcomes

Other covariates are found to be significantly associated with WHZ/wasting but not with HAZ/stunting. We speculate that the underlying determinants of nutrition, such as wealth, education, maternal care, and disease and sanitation, are significantly attenuated for HAZ/stunting when younger children (6–23 months) are combined with older children (24–59 months). With WHZ/wasting, the opposite effect is observed due to the inclusion of older children (Alderman and Headey, 2018). In this study, we did not separate by age group due to the small sample size.

HAZ

Only Workload is positively associated with HAZ (Table 2, Panel 7). The association is robust across different estimation techniques. It especially impacts girls¹² (1.604 SD increase, p < 0.05), a finding similar to previous studies (Ruel et al., 1999; Abate and Belachew, 2017). Most rural households operate as a family farming system, wherein farm operation is primarily dependent on the family's manual labor and the use of animals (Tran, 2003; Ye and Pan, 2016). To balance their agricultural obligations and their maternal and caring roles, mothers carry their babies on their backs while working in the fields. Further research is required to tease out the quality of childcare practices in the target locations because our results indicate that, in addition to income gains from modern market participation, maternal care also significantly impacts the linear growth of girls and their future offspring (Walker et al., 2015).

Stunting

Maternal height (HtMother) and the geographic stratum HV-HA are important covariates for stunting. HtMother is associated with a lower probability of stunting (p < 0.05), although the effect size is small. The association is true for both boys and girls supporting findings of other studies (e.g., Prendergast and Humphrey, 2014), which found association between stunted mothers and their children having increased likelihood of stunting. Maternal height did not have a significant effect on children's HAZ, which could suggest that other external factors, aside from the genetic predisposition, come into play; for instance, if women of reproductive age have access to adequate health and nutrition, and health care (De Onis and Branca, 2016). Boys in HV-HA have lower prevalence of stunting compared to those in the LV-LA areas¹³. This result shows that higher altitude does not imply higher incidence of stunting, and higher vegetable density per capita may have additional advantages compared with lower vegetable density per capita. Predictors that magnify the probability of stunting are AgeMother and BMIMother (Table 4, Panel 7). However, the effect sizes are small and the association (BMIMother) is marginal.

WHZ

Overall, EducMother (p < 0.05) and MongMother (p < 0.10) are positively associated with WHZ. Further, for both boys and girls, NonFoodExp_pc is also positively associated with WHZ, indicating that household income matters (Martorell and Zongrone, 2012). For boys, WHZ improves as they get older (AgeChild, p < 0.05). These results are robust, but the effect sizes are small. BMIMother also positively affects boys' WHZ, although marginally. Considering geographic location, boys in the HV-HA stratum are associated with lower WHZ (p < 0.10) than those in the LV-LA stratum. For girls, Area, another indication of household wealth, positively but marginally affects WHZ (p < 0.10), while Income, Workload and Diarrhea negatively affect WHZ. These relationships are at the 1- and 5-percent levels of significance. The negative association between Diarrhea and WHZ is especially robust in different estimation techniques. However, the impact of the two women's empowerment variables (Income and Workload) on WHZ is negative and highly significant at the 1- and 5-percent levels. Nonetheless, while Workload negatively affects WHZ (short-term undernutrition), it has a larger positive effect on HAZ (long-term undernutrition).

Wasting

The results resonate with our findings in WHZ, where maternal education (EducMother) and ethnicity (MongMother) positively reduce the probability of wasting, in addition to Income and AgeChild (p < 0.05). However, the effect sizes of AgeChild and EducMother are relatively small. Among boys, the probability of wasting is reduced via Income, AgeChild and BMIMother. The probability of wasting among girls is strongly reduced via NonFoodExp_pc (p < 0.05). Diarrhea, as previously shown, increases the probability of wasting, especially among girls, as does Workload. For Workload, reduced workload means more time spent at home and less time spent helping in productive agricultural activities. This may negatively affect the household's meager disposable income, and therefore alter the amount, quantity, composition and quality of food purchased and consumed by these children (Carletto et al., 2015). The prevalence is increased by LV-HA and HV-HA areas in relation to LV-LA, which indicates that elevation may affect the increased incidence of wasting among boys. As previously mentioned, these results for boys require further validation due to the small number of observations.

These results indicate that wealth (NonFoodExp_pc) and maternal education are important for addressing short-term undernutrition consistent with other studies (Frost et al., 2005; Martorell and Zongrone, 2012; Makate and Makate, 2017). However, addressing short-term undernutrition differs for boys and girls due to preferential treatment given to sons among these ethnic minority communities (Jones et al., 2014). Mothers allocate more health-promoting resources and childcare time to sons, thereby negatively affecting girls' nutritional status in the short-term (Barcellos et al., 2014). This might also explain the positive, but marginal relationship, between maternal BMI and boy's WHZ. The incidence of diarrhea severely affects girls' WHZ/wasting, and thus requires future research to identify its determinants.

Summary, Discussion, and Limitations

This cross-sectional study analyzed 234 children aged 6–60 months and has shown how smallholder vegetable production may lead to positive nutritional outcomes through improving a child's linear growth (HAZ) and weight (WHZ and WAZ), and reducing the prevalence of stunting, wasting, and underweight. To our knowledge, this is the first study that assesses the contribution of smallholder vegetable production on child nutrition among ethnic minority communities. Our results are broadly consistent with the results of other studies showing how agriculture in general, and specific crop/livestock in particular, can lead to improved children's nutrition (e.g., Hoddinott et al., 2015; Kumar et al., 2015; Malapit et al., 2015; Shively and Sununtnasuk, 2015). However, the impact on children's nutrition varies for each of the smallholder vegetable production indicators and by gender, controlling for other covariates in the econometric models.

About 80 percent of smallholder households in the province mainly grow vegetables for home consumption. For these subsistence farmers, diversifying their vegetable production has an added advantage. It has benefits in that it improves boys' WHZ and girls' HAZ. However, its health benefits are mixed because it leads to a higher prevalence of stunting and underweight among boys. It may significantly improve girls' HAZ, but no significant association has been found to mitigate the stunting outcome.

Our results suggest that smallholder vegetable production has a significant indirect effect on children's nutrition via market participation. Among boys, market participation is associated with a lower probability of stunting and underweight. Our results also indicate that other covariates, such as maternal education and diarrhea incidence, are significantly associated with the short-term nutritional outcomes (WHZ and wasting).

What do these results imply for smallholder vegetable producers in Lao Cai province, Vietnam? Our results point at issues that require further study to determine whether there is causal relationship between these variables, and not just a correlation. If that is the case, there is a need to create and/or improve existing market linkages between smallholder vegetable producers and markets. The type of linkage will depend on the specific conditions in each locality: farmer-to-trader linkage, farmer-to-retailer linkage, or creation of farmer groups or cooperatives. While markets present opportunities for rural farmers to increase their livelihoods, it can also lead to higher transaction costs if they are not suitably adapted to the locality, thus inducing some households to opt out of market participation (Fackler and Goodwin, 2001; Renkow et al., 2004). Road infrastructure investments or better information about market opportunities and prices, can also likely improve children's nutritional outcomes in the long-term, particularly with respect to reduced stunting.

In addition to providing income-generating opportunities for these households, future interventions could also incorporate the following factors that are strongly associated with child nutritional outcomes. The first is to promote education among the ethnic minority communities because higher maternal educational attainment increases the allocative efficiency of health production. Maternal education is a significant predictor of a child's disease incidence, as well as the ability to report or seek medical care at community health centers when their children are suffering from diarrhea (Rheinlander et al., 2010). The odds of a child being reported slightly sick is 25 percent higher if the mother has some primary schooling rather than being illiterate (Teerawichitchainan and Phillips, 2008). Education also provides access to more information on balanced diets and nutrition, which translates to better nutritional knowledge and better feeding practices (Makate and Makate, 2017), and therefore higher demand for diverse foods (Moon et al., 2002) and healthy food choices (Variyam et al., 1998). One study in Vietnam (Nguyen et al., 2013) found that children whose mothers had reached high school education were 1.7 times more likely to achieve the minimum dietary diversity than those mothers who did not go to school. Among Mong food preparers, improving their nutritional knowledge, by providing them information about healthy eating, could be included in the design of any future health-promoting intervention. Young children from ethnic minority communities (Thai-Muong, Tay-Nung, E De-Mnong) suffer lower dietary diversity compared to the children of Kinh mothers since ethnic minority mothers give their young children fewer legumes and nuts, dairy products, flesh foods, and vitamin-A rich fruits and vegetables (Nguyen et al., 2016). While the aim is not to impose change on their ethnic heritage, nutritional education could highlight the benefits of diversified diets, which can translate to better child nutritional outcomes. Nutritional education can increase children's dietary diversity, as shown by previous studies in Cambodia, Malawi, and Western Kenya (Waswa et al., 2015; Reinbott et al., 2016; Kuchenbecker et al., 2017). When dietary diversity is promoted through nutritional education, it can be easily integrated into the local knowledge (Powell et al., 2017). Further research on the quality of maternal care-giving practices is also encouraged because it has important implications for the improvement of girls' HAZ, WHZ, and wasting.

Furthermore, the country has had numerous policy actions aimed at improving safe water supply and constructing sanitation facilities in rural areas. However, uptake of the water and sanitation facilities among ethnic minority communities has been slow since the current hygiene and sanitation policies fail to understand the local cultural and social factors that guide their sanitation behaviors (Rheinlander et al., 2010). Therefore, sanitation solutions that are in line with the local priorities and hygiene perceptions, and are designed appropriately to suit their living conditions, could ensure the effectiveness of future hygiene-promotion strategies (Rheinlander et al., 2010).

The observed gender differences with respect to health inequality in our sample may be due to either preferential treatment toward one gender or be biological in nature. In Sub-Saharan Africa, the preferential treatment of girls gives rise to their slight anthropometric advantage over boys (Svedberg, 1990). However, biologically, boys are more vulnerable to stunting than girls (Wamani et al., 2007). There may be other contributing factor(s) to this that our survey was not able to capture.

In the broader context, the above results highlight similar findings of previous studies in Vietnam where significant reductions in child undernutrition can be driven by improvements in household income, maternal education, hygiene, and health and nutrition services (Kien et al., 2016; Nguyen et al., 2017; Beal et al., 2019; Harris et al., 2021). Research has shown that hard infrastructure alone, e.g., provision of roads or sanitation facilities, must be coupled with adequate levels of soft infrastructure investments (human capital and innovation) to maximize impact on the target beneficiaries. This is crucial given the deeper structural issues underpinning nutritional inequities among ethnic minority communities in Vietnam. For instance, how do we encourage the ethnic minorities to adopt vegetable diversification or participate in markets while taking into account their resistance to state intervention in favor of their traditional indigenous knowledge either due to cultural and language barriers, mistrust due to state territorialization and fragmentation of their traditional social geography, or that the changes brought about by these interventions are too drastic and undermine their traditional way of life (Bonnin and Turner, 2012; Harris et al., 2021)?

Currently, their agricultural productivity remains low and they are resistant to change. They use local varieties with limited inputs that result in lower yields. They also tend to operate on subsistence farming and lack the entrepreneurial impetus, with mismatched interests and abilities on the production of high-value industrial crops. In terms of market engagement, they use their cultural networks and avoid the Kinh-dominated markets (Minh, 2019; World Bank, 2019). All these may have to be taken into account when tackling child undernutrition.

An enabling environment may also be needed to support these socioeconomic drivers. Harris et al. (2021) has identified three crucial areas to achieve stunting reduction in ethnic minority communities. These are: (a) adaptation of a differentiated socioeconomic policy for ethnic minorities that highlight the specific needs and preferences of these communities; (b) prioritization of nutrition funding and capacity building for provinces with high minority populations to overcome the barriers for service delivery; and (c) engagement of more national and international civil societies that champion the inclusion of ethnic minorities into the nutrition policymaking process.

This study has several limitations that need to be considered. Firstly, it is based on cross-sectional data that inhibits us to capture variations in household behavior over time. Secondly, the sampled households are from districts with high geographic concentrations of ethnic minority communities and are thus not comparable to previous child undernutrition studies using the Vietnam Nutrition Surveillance System (Beal et al., 2019) or Vietnam Multiple Indicator Cluster Survey (Kien et al., 2016; Harris et al., 2021), where ethnic minorities are underrepresented. Also, other variables, like child immunization, vitamin A supplementation, maternal breastfeeding practices, or economic shocks, are not included in the analysis due to data limitations. The small number of observations also prevents us from conducting disaggregated analysis by age group, which would have been more meaningful since various underlying determinants of children's nutrition outcomes are age-sensitive (Alderman and Headey, 2018). Despite these limitations, these results do show that improved child nutrition outcomes in Lao Cai province can be achieved by improving livelihoods via smallholder vegetable production and market participation. Further research is encouraged to ascertain the causal relationships using more observations and longitudinal or panel data.

In conclusion, child undernutrition continues to be a public health concern in Lao Cai province, where stunting rate remains one of the highest in the country. This study presents evidence that smallholder vegetable production can be an alternative and sustainable solution to address child undernutrition, especially among ethnic minority communities in remote areas. It can be replicated in other geographic locations with high concentrations of disadvantaged groups to assist in the design and implementation of nutritional and health promotion policies using the available resource as part of the solution.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

The studies involving human participants were reviewed and approved by the Low Risk Human Research Ethics Review Group (Faculty of the Arts and Faculty of the Professions), Compliance and Integrity, Research Services, The University of Adelaide (H-2015-159). Written informed consent was obtained from all individual respondents in the study.

Author Contributions

CG involved in the study design, performed primary data collection, data encoding, data cleaning, data analysis, data interpretation, and wrote and revised the manuscript. WU guided study design and data collection, supervised data analysis and data interpretation, provided intellectual input, and edited the manuscript. AP guided study design and data collection, supervised data analysis and data interpretation, and edited the manuscript. SN guided study design and data collection, contributed in data interpretation, and edited the manuscript. DZ supervised data analysis and data interpretation and edited the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by the Australian Center for International Agricultural Research (ACIAR) through AGB/2012/059.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Acknowledgments

We acknowledge the financial support by the Australian Center for International Agricultural Research (ACIAR) through AGB/2012/059. We are also immensely grateful to the Vietnam Women's Union (VWU), Lao Cai Department of Crop Production and Crop Protection, Mekong Development Research Institute (MDRI), Vietnam National University of Agriculture (VNUA), and Oikoi (formerly Agricultural Impact International), for their assistance during the primary data collection. We also acknowledge the support from the local officials and rural households who have accommodated us during our research in Lao Cai province, Vietnam. Lastly, we would like to thank the two anonymous reviewers for their suggestions during the peer review process, which improved the quality of the article.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsufs.2022.900625/full#supplementary-material

Footnotes

1. ^We purposely included one commune and three villages in Sa Pa district to be of some benefit to the project which funded this research.

2. ^MDRI is a research institute in Hanoi, Vietnam that was contracted to assist our team in gathering data in the northwest.

3. ^https://www.dimagi.com/commcare/

4. ^A tape measure is a retractable flexible ruler that is used to measure length or distance.

5. ^The food groups in the 1994 and the 2007 versions of the Vietnamese Food Composition Tables are similar, except that the ordering of food group “softdrink, beverages, alcoholic beverages (13)” and “condiments and traditional sauces (14)” are interchanged.

6. ^L=low, H=high, A=altitude/elevation, V=vegetable per capita density.

7. ^Improved toilet facilities include: septic tank, biogas, double pit dry, ventilated pit dry, and single pit dry; unimproved toilets include no toilet, pour flush, fishpond, and ashes/bridge/bucket (World Health Organization and United Nations Children's Fund, 2006).

8. ^Instead of the Durbin Wu-Hausman statistics for endogeneity test, Wooldridge's (1995) robust score test and robust regression-based test statistics are reported after 2SLS estimation with a robust VCE (Wooldridge, 1995 as cited by StataCorp, 2017).

9. ^As part of a system of equations, we use distance to an agricultural office (Distance) as an instrument for VegDiversity. The use of Distance could be a good instrument of VegDiversity, since proximity to an agricultural extension office, which can be found in most communes, means easy access to information on improved crop production practices and free seeds, and has therefore the potential to affect VegDiversity (Di Falco et al., 2011).

10. ^In the regression model, education is treated as a continuous variable rather than a categorical variable to determine the effect of higher education on the child nutritional outcomes.

11. ^1 US Dollar (US$) = 23,208.37 Vietnamese Dong (VND) (2020 average). Source: UNCTAD, ‘Currency exchange rates, annual, 1970-2020', http://unctadstat.unctad.org/wds/TableViewer/tableView.aspx.

12. ^Full models with all control variables used are in the supplemental online materials (Supplementary Tables 3A,B).

13. ^Full models with all control variables used are in the supplemental online materials (Supplementary Tables 5A,B).

References

Abate, K. H., and Belachew, T. (2017). Care and not wealth is a predictor of wasting and stunting of 'The Coffee Kids' of Jimma Zone, Southwest Ethiopia. Nutr. Health 23, 193–202. doi: 10.1177/0260106017706253

PubMed Abstract | CrossRef Full Text | Google Scholar

Abay, K., and Hirvonen, K. (2016). Does market access mitigate the impact of seasonality on child growth? Panel data evidence from Northern Ethiopia. J. Dev. Stud. 53, 1414–1429. doi: 10.1080/00220388.2016.1251586

CrossRef Full Text | Google Scholar

Abdullah, M., and Wheeler, E. F. (1985). Seasonal variations, and the intra-household distribution of food in a Bangladeshi village. Am. J. Clin. Nutr. 41, 1305–1313. doi: 10.1093/ajcn/41.6.1305

PubMed Abstract | CrossRef Full Text | Google Scholar

Addo, O. Y., Stein, A. D., Fall, C. H., Gigante, D. P., Guntupalli, A. M., Horta, B. L., et al. (2013). Maternal height and child growth patterns. J. Pediatr. 163, 549–554.e1. doi: 10.1016/j.jpeds.2013.02.002

PubMed Abstract | CrossRef Full Text | Google Scholar

Alderman, H., Elder, L., Goyal, A., Herforth, A., Hoberg, Y. T., Marini, A., et al. (2013). Improving Nutrition Through Multisectoral Approaches. Washington, DC: The World Bank.

Alderman, H., and Headey, D. (2017). How important is parental education for child nutrition? World Dev. 94, 448–464. doi: 10.1016/j.worlddev.2017.02.007

PubMed Abstract | CrossRef Full Text | Google Scholar

Alderman, H., and Headey, D. (2018). The timing of growth faltering has important implications for observational analyses of the underlying determinants of nutrition outcomes. PLoS ONE 13, e0195904. doi: 10.1371/journal.pone.0195904

PubMed Abstract | CrossRef Full Text | Google Scholar

Andersson, C. I. M., Chege, C. G. K., Rao, E. J. O., and Qaim, M. (2015). Following up on smallholder farmers and supermarkets in Kenya. Am. J. Agric. Econ. 97, 1247–1266. doi: 10.1093/ajae/aav006

CrossRef Full Text | Google Scholar

Angrist, J. D., and Pischke, J. S. (2009). Mostly Harmless Econometrics: An Empiricist's Companion. Princeton, NJ: Princeton University Press. doi: 10.1515/9781400829828

CrossRef Full Text | Google Scholar

Association of Southeast Asian Nations (ASEAN), United Nations Children's Fund (UNICEF), and World Health Organization (WHO). (2016). Regional Report on Nutrition Security in ASEAN Volume 2. Bangkok: UNICEF.

Babu, S. C., Gajanan, S. N., and Sanyal, P. (2014). “Household care as a determinant of nutritional status— Application of instrumental variable estimation,” in Food Security, Poverty and Nutrition Policy Analysis, eds S. C. Babu, S. N. Gajanan. and P. Sanyal (San Diego, CA: Academic Press), 441–460. doi: 10.1016/B978-0-12-405864-4.00014-4

CrossRef Full Text | Google Scholar

Barcellos, S. H., Carvalho, L. S., and Lleras-Muney, A. (2014). Child gender and parental investments in India: are boys and girls treated differently? Am. Econ. J. Appl. Econ. 6, 157–189. doi: 10.1257/app.6.1.157

PubMed Abstract | CrossRef Full Text | Google Scholar

Beal, T., Le, D. T., Trinh, T. H., Burra, D. D., Huynh, T., Duong, T. T., et al. (2019). Child stunting is associated with child, maternal, and environmental factors in Vietnam. Matern. Child Nutr. 15, e12826. doi: 10.1111/mcn.12826

PubMed Abstract | CrossRef Full Text | Google Scholar

Black, R. E., Victora, C. G., Walker, S. P., Bhutta, Z. A., Christian, P., de Onis, M., et al. (2013). Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 382, 427–451. doi: 10.1016/S0140-6736(13)60937-X

PubMed Abstract | CrossRef Full Text | Google Scholar

Bonnin, C., and Turner, S. (2012). Livelihood vulnerability and food security among upland ethnic minorities in Northern Vietnam. Kasarinlan Philippine J. Third World Stud. 26, 324–340. Retrieved from: https://journals.upd.edu.ph/index.php/kasarinlan/article/view/3503 (accessed July 25, 2018).

Carletto, G., Ruel, M., Winters, P., and Zezza, A. (2015). Farm-level pathways to improved nutritional status: introduction to the special issue. J. Dev. Stud. 51, 945–957. doi: 10.1080/00220388.2015.1018908

CrossRef Full Text | Google Scholar

Centers for Disease Control and Prevention (CDC) (2016). National Health and Nutritional Examination Survey (NHANES) Anthropometry Procedures Manual. Retrieved from: https://www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf (accessed July 25, 2008).

Chaparro, C., Oot, L., and Sethuraman, K. (2014). Vietnam Nutrition Profile. Washington, DC: FHI 360/FANTA.

Cunningham, K., Ferguson, E., Ruel, M., Uauy, R., Kadiyala, S., Menon, P., et al. (2019). Water, sanitation, and hygiene practices mediate the association between women's empowerment and child length-for-age z-scores in Nepal. Matern. Child Nutr. 15, e12638. doi: 10.1111/mcn.12638

PubMed Abstract | CrossRef Full Text | Google Scholar

Dang, H.-A. (2012). “Vietnam: a widening poverty gap for ethnic minorities,” in Indigenous Peoples, Poverty, and Development, eds G. Hall and H. A. Patrinos (New York, NY: Cambridge University Press), 304–343. doi: 10.1017/CBO9781139105729.008

CrossRef Full Text | Google Scholar

Dangour, A. D., Green, R., Hasler, B., Rushton, J., Shankar, B., and Waage, J. (2012). Linking agriculture and health in low- and middle-income countries: an interdisciplinary research agenda. Proc. Nutr. Soc. 71, 222–228. doi: 10.1017/S0029665112000213

PubMed Abstract | CrossRef Full Text | Google Scholar

De Onis, M., and Branca, F. (2016). Childhood stunting: a global perspective. Matern. Child Nutr. 12, 12–26. doi: 10.1111/mcn.12231

PubMed Abstract | CrossRef Full Text | Google Scholar

Di Falco, S., Veronesi, M., and Yesuf, M. (2011). Does adaptation to climate change provide food security? A micro-perspective from Ethiopia. Am. J. Agric. Econ. 93, 829–846. doi: 10.1093/ajae/aar006

CrossRef Full Text | Google Scholar

Epprecht, M., Müller, D., and Minot, N. (2009). How remote are Vietnam's ethnic minorities? An analysis of spatial patterns of poverty and inequality. Ann. Regional Sci. 46, 349–368. doi: 10.1007/s00168-009-0330-7

CrossRef Full Text | Google Scholar

Fackler, P. L., and Goodwin, B. K. (2001). “Chapter 17 Spatial price analysis,” in Marketing, Distribution and Consumers, eds B. L. Gardner and G. C. Rausser (Amsterdam: Elsevier Science B.V), 971–1024. doi: 10.1016/S1574-0072(01)10025-3

CrossRef Full Text | Google Scholar

Food and Agriculture Organization Corporate Statistical Database (FAOSTAT) (2022). Vietnam [Online data on 2019 vegetable primary total]. Retrieved from: https://www.fao.org/faostat/en/#data/QCL (accessed March 10, 2008).

Forero-Ramirez, N., Gamboa, L. F., Bedi, A., and Sparrow, R. (2014). Desnutricion infantil y atencion prenatal: datos probatorios de tres paises latinoamericanos [Child malnutrition and prenatal care: evidence from three Latin American countries]. Revista Panamericana de Salud Publica 35, 163–171. Retrieved from: https://www.scielosp.org/article/rpsp/2014.v35n3/163-171/ (accessed July 25, 2018).

Frost, M. B., Forste, R., and Haas, D. W. (2005). Maternal education and child nutritional status in Bolivia: finding the links. Soc. Sci. Med. 60, 395–407. doi: 10.1016/j.socscimed.2004.05.010

PubMed Abstract | CrossRef Full Text | Google Scholar

General Statistics Office (GSO) of Vietnam (2020). Completed Results of the 2019 Viet Nam Population and Housing Census. Hanoi: Statistical Publishing House.

Genova, C., Umberger, W., Newman, S., and Peralta, A. (2022). Understanding food choice factors of rural households from northwest Vietnam. J. Agribusiness Dev. Emerg. Econ. doi: 10.1108/JADEE-12-2021-0320

CrossRef Full Text | Google Scholar

Gillespie, S., Harris, J., and Kadiyala, S. (2012). The Agriculture-Nutrition Disconnect in India: What Do We Know (IFPRI Discussion Paper 01187). Washington, DC: International Food Policy Research Institute.

Gittelsohn, J. (1991). Opening the box: intrahousehold food allocation in rural Nepal. Soc. Sci. Med. 33, 1141–1154. doi: 10.1016/0277-9536(91)90230-A

PubMed Abstract | CrossRef Full Text | Google Scholar

Global Nutrition Report (2022). Country Nutrition Profiles - Vietnam. Bristol: Development Initiatives. Retrieved from: https://globalnutritionreport.org/resources/nutrition-profiles/asia/south-eastern-asia/viet-nam/#overview (accessed May 30, 2022).

Goli, S., Rammohan, A., and Singh, D. (2015). The effect of early marriages and early childbearing on women's nutritional status in India. Matern. Child Health J. 19, 1864–1880. doi: 10.1007/s10995-015-1700-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Harris, J., Huynh, P., Nguyen, H. T., Hoang, N., Mai, L. T., Tuyen, L. D., et al. (2021). Nobody left behind? Equity and the drivers of stunting reduction in Vietnamese ethnic minority populations. Food Security 13, 803–818. doi: 10.1007/s12571-021-01183-7

CrossRef Full Text | Google Scholar

Headey, D., Chiu, A., and Kadiyala, S. (2012). Agriculture's role in the Indian enigma: help or hindrance to the crisis of undernutrition? Food Security 4, 87–102. doi: 10.1007/s12571-011-0161-0

CrossRef Full Text | Google Scholar

Heckert, J., Olney, D. K., and Ruel, M. T. (2019). Is women's empowerment a pathway to improving child nutrition outcomes in a nutrition-sensitive agriculture program?: Evidence from a randomized controlled trial in Burkina Faso. Soc. Sci. Med. 233, 93–102. doi: 10.1016/j.socscimed.2019.05.016

PubMed Abstract | CrossRef Full Text | Google Scholar

Hellevik, O. (2009). Linear versus logistic regression when the dependent variable is a dichotomy. Qual. Quant. 43, 59–74. doi: 10.1007/s11135-007-9077-3

CrossRef Full Text | Google Scholar

Herforth, A., and Ballard, T. J. (2016). Nutrition indicators in agriculture projects: current measurement, priorities, and gaps. Glob, Food Security-Agric. Policy Econ. Environ. 10, 1–10. doi: 10.1016/j.gfs.2016.07.004

CrossRef Full Text | Google Scholar

Herforth, A., and Harris, J. (2014). Understanding and Applying Primary Pathways and Principles (Improving Nutrition through Agriculture Technical Brief Series). Arlington, VA: USAID/Strengthening Partnerships, Results, and Innovations in Nutrition Globally (SPRING) Project.

Hirvonen, K., and Hoddinott, J. (2017). Agricultural production and children's diets: evidence from rural Ethiopia. Agric. Econ. 48, 469–480. doi: 10.1111/agec.12348

CrossRef Full Text | Google Scholar

Hoddinott, J., Headey, D., and Dereje, M. (2015). Cows, missing milk markets, and nutrition in rural Ethiopia. J. Dev. Stud. 51, 958–975. doi: 10.1080/00220388.2015.1018903

CrossRef Full Text | Google Scholar

IFPRI (2016). Global Nutrition Report 2016: From Promise to Impact: Ending Malnutrition by 2030. Washington, DC: IFPRI.

International Food Policy Research Institute (IFPRI) (2002). Fruits and vegetables in Vietnam: Adding value from farmer to consumer. Washington, DC: International Food Policy Research Institute (IFPRI).

International Livestock Research Institute (ILRI) (2014). A Situational Analysis of Agricultural Production and Marketing, and Natural Resources Management Systems in Northwest Vietnam. Nairobi: International Livestock Research Institute for CGIAR Humidtropics Research Program.

Jones, N., Presler-Marshall, E., and Anh, T. T. V. (2014). Early Marriage Among Viet Nam's Hmong: How Unevenly Changing Gender Norms Limit Hmong Adolescent Girls' Options in Marriage and Life. London: Overseas Development Institute.

Kadiyala, S., Harris, J., Headey, D., Yosef, S., and Gillespie, S. (2014). Agriculture and nutrition in India: mapping evidence to pathways. Ann. N. Y. Acad. Sci. 1331, 43–56. doi: 10.1111/nyas.12477

PubMed Abstract | CrossRef Full Text | Google Scholar

Kanter, R., Walls, H. L., Tak, M., Roberts, F., and Waage, J. (2015). A conceptual framework for understanding the impacts of agriculture and food system policies on nutrition and health. Food Security 7, 767–777. doi: 10.1007/s12571-015-0473-6

CrossRef Full Text | Google Scholar

Keding, G. B., Msuya, J. M., Maass, B. L., and Krawinkel, M. B. (2012). Relating dietary diversity and food variety scores to vegetable production and socio-economic status of women in rural Tanzania. Food Security 4, 129–140. doi: 10.1007/s12571-011-0163-y

CrossRef Full Text | Google Scholar

Kien, V. D., Lee, H. Y., Nam, Y. S., Oh, J., Giang, K. B., and Van Minh, H. (2016). Trends in socioeconomic inequalities in child malnutrition in Vietnam: Findings from the Multiple Indicator Cluster Surveys, 2000-2011. Glob. Health Action 9, 29263. doi: 10.3402/gha.v9.29263

PubMed Abstract | CrossRef Full Text | Google Scholar

Kim, C., ALvarez, C., Sattar, A., Bandyopadhyay, A., Azzarri, C., Moltedo, A., et al. (2021). Production, consumption, and food security in Viet Nam: Diagnostic overview. Washington, DC: IFPRI and FAO.

Kimenju, S. C., and Qaim, M. (2016). The nutrition transition and indicators of child malnutrition. Food Security 8, 571–583. doi: 10.1007/s12571-016-0566-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Koppmair, S., Kassie, M., and Qaim, M. (2017). Farm production, market access and dietary diversity in Malawi. Public Health Nutr. 20, 325–335. doi: 10.1017/S1368980016002135

PubMed Abstract | CrossRef Full Text | Google Scholar

Kozel, V. (2014). “Reducing poverty among ethnic minorities,” in Well Begun but not yet Done: Progress and Emerging Challenges for Poverty Reduction in Vietnam, ed V. Kozel (Washington, DC: World Bank), 163–193. doi: 10.1596/978-1-4648-0006-1_ch5

CrossRef Full Text | Google Scholar

Kuchenbecker, J., Reinbott, A., Mtimuni, B., Krawinkel, M. B., and Jordan, I. (2017). Nutrition education improves dietary diversity of children 6-23 months at community-level: Results from a cluster randomized controlled trial in Malawi. PLoS ONE 12, e0175216. doi: 10.1371/journal.pone.0175216

PubMed Abstract | CrossRef Full Text | Google Scholar

Kumar, N., Harris, J., and Rawat, R. (2015). If they grow it, will they eat and grow? Evidence from Zambia on agricultural diversity and child undernutrition. J. Dev. Stud. 51, 1060–1077. doi: 10.1080/00220388.2015.1018901

CrossRef Full Text | Google Scholar

Lao Cai People's Committee (2016). Dân số [Population]. Retrieved from: http://www.laocai.gov.vn/ubnd-laocai/4/469/38156/250321/Gioi-thieu-ve-tinh-Lao-Cai-/Dan-so.aspx (accessed August 5, 2016).

Leroy, J. L. (2011). zscore06: Stata Module to Calculate Anthropometric Z-Scores Using the 2006 WHO Child Growth Standards (Statistical Software Components S4572279]. Retrieved from: https://ideas.repec.org/c/boc/bocode/s457279.html (accessed July 25, 2018).

Linh, V., and Glewwe, P. (2011). Impacts of rising food prices on poverty and welfare in Vietnam. J. Agric. Resour. Econ. 36, 14–27. Retrieved from: https://www.proquest.com/docview/1114146595/fulltextPDF/56D2968F1A1E4D3FPQ/1?accountid=26874 (accessed July 25, 2018).

Luo, W., Zhai, F., Jin, S., and Ge, K. (2001). Intrahousehold food distribution: a case study of eight provinces in China. Asia Pac. J. Clin. Nutr. 10(Suppl), S19–28. doi: 10.1046/j.1440-6047.2001.0100s1S19.x

PubMed Abstract | CrossRef Full Text | Google Scholar

Madjdian, D. S., and Bras, H. A. J. (2016). Family, gender, and women's nutritional status: a comparison between two Himalayan communities in Nepal. Econ. History Dev. Regions 31, 198–223. doi: 10.1080/20780389.2015.1114416

CrossRef Full Text | Google Scholar

Makate, M., and Makate, C. (2017). Educated mothers, well-fed and healthy children? Assessing the impact of the 1980 school reform on dietary diversity and nutrition outcomes of Zimbabwean children. J. Dev. Stud. 54, 1196–1216. doi: 10.1080/00220388.2017.1380796

CrossRef Full Text | Google Scholar

Malapit, H., Kadiyala, S., Quisumbing, A., Cunningham, K., and Tyagi, P. (2015). Women's empowerment mitigates the negative effects of low production diversity on maternal and child nutrition in Nepal. J. Dev. Stud. 51, 1097–1123. doi: 10.1080/00220388.2015.1018904

CrossRef Full Text | Google Scholar

Malapit, H., Pinkstaff, C., Sproule, K., Kovarik, C., Quisumbing, A., and Meinzen-Dick, R. (2017). The Abbreviated Women's Empowerment in Agriculture Index (A-WEAI) (IFPRI Discussion Paper 01647). Washington, DC: International Food Policy Research Institute.

Martorell, R., and Zongrone, A. (2012). Intergenerational influences on child growth and undernutrition. Paediatr. Perinat. Epidemiol. 26, 302–314. doi: 10.1111/j.1365-3016.2012.01298.x

PubMed Abstract | CrossRef Full Text | Google Scholar

Masset, E., Haddad, L., Cornelius, A., and Isaza-Castro, J. (2012). Effectiveness of agricultural interventions that aim to improve nutritional status of children: systematic review. BMJ 344, d8222. doi: 10.1136/bmj.d8222

PubMed Abstract | CrossRef Full Text | Google Scholar

Minh, T. T. (2019). Unpacking the systemic problems and blocking mechanisms of a regional agricultural innovation system: an integrated regional-functional-structural analysis. Agric. Syst. 173, 268–280. doi: 10.1016/j.agsy.2019.03.009

CrossRef Full Text | Google Scholar

Moon, W., Florkowski, W. J., Beuchat, L. R., Resurreccion, A. V., Paraskova, P., Jordanov, J., et al. (2002). Demand for food variety in an emerging market economy. Appl. Econ. 34, 573–581. doi: 10.1080/00036840110037863

CrossRef Full Text | Google Scholar

Muller, C. (2009). Do agricultural outputs of partly autarkic peasants affect their health and nutrition? Evidence from Rwanda. Food Policy 34, 166–175. doi: 10.1016/j.foodpol.2008.10.010

CrossRef Full Text | Google Scholar

Murphy, R., Tao, R., and Lu, X. (2011). Son preference in rural China: patrilineal families and socioeconomic change. Popul. Dev. Rev. 37, 665–690. doi: 10.1111/j.1728-4457.2011.00452.x

PubMed Abstract | CrossRef Full Text | Google Scholar

National Institute of Nutrition (2019). Prevalence of Undernutrition by Severity - by 6 Ecological Region 2019. Retrieved from: http://chuyentrang.viendinhduong.vn/FileUpload/Documents/Nam%202021/Ty%20le%20SDD%20tre%20em%20nam%202019.pdf (accessed May 30, 2022).

National Institute of Nutrition and United Nations Children's Fund (2011). A Review of the Nutrition Situation in Vietnam 2009-2010. Hanoi: Medical Publishing House.

Nguyen, P. H., Headey, D., Frongillo, E. A., Tran, L. M., Rawat, R., Ruel, M. T., et al. (2017). Changes in underlying determinants explain rapid increases in child linear growth in Alive & Thrive study areas between 2010 and 2014 in Bangladesh and Vietnam. J. Nutr. 147, 462–469. doi: 10.3945/jn.116.243949

PubMed Abstract | CrossRef Full Text | Google Scholar

Nguyen, P. H., Strizich, G., Lowe, A., Nguyen, H., Pham, H., Truong, T. V., et al. (2013). Food consumption patterns and associated factors among Vietnamese women of reproductive age. Nutr. J. 12, 126. doi: 10.1186/1475-2891-12-126

PubMed Abstract | CrossRef Full Text | Google Scholar

Nguyen, T. T., Nguyen, P. H., Hajeebhoy, N., Nguyen, H. V., and Frongillo, E. A. (2016). Infant and young child feeding practices differ by ethnicity of Vietnamese mothers. BMC Pregnancy Childbirth 16, 214. doi: 10.1186/s12884-016-0995-8

PubMed Abstract | CrossRef Full Text | Google Scholar

O'donnell, O., Van Doorslaer, E., Wagstaff, A., and Lindelow, M. (2008). Analyzing Health Equity Using Household Survey Data: A Guide to Techniques and Their Implementation. Washington, DC: The World Bank. doi: 10.1596/978-0-8213-6933-3

CrossRef Full Text | Google Scholar

Parsons, J., Edmeades, J., Kes, A., Petroni, S., Sexton, M., and Wodon, Q. (2015). Economic impacts of child marriage: a review of the literature. Rev. Faith Int. Affairs 13, 12–22. doi: 10.1080/15570274.2015.1075757

CrossRef Full Text | Google Scholar

Pinstrup-Andersen, P. (2014). “Food systems and human nutrition: relationships and policy interventions,” in Improving Diets and Nutrition: Food-Based Approaches, eds B. Thompson and L. Amoroso (Rome: CABI and FAO), 8–18. doi: 10.1079/9781780642994.0008

CrossRef Full Text | Google Scholar

Powell, B., Bezner Kerr, R., Young, S. L., and Johns, T. (2017). The determinants of dietary diversity and nutrition: Ethnonutrition knowledge of local people in the East Usambara Mountains, Tanzania. J. Ethnobiol. Ethnomed. 13, 23. doi: 10.1186/s13002-017-0150-2

PubMed Abstract | CrossRef Full Text | Google Scholar

Prendergast, A. J., and Humphrey, J. H. (2014). The stunting syndrome in developing countries. Paediatr. Int. Child Health 34, 250–265. doi: 10.1179/2046905514Y.0000000158

PubMed Abstract | CrossRef Full Text | Google Scholar

Quisumbing, A. R., Brown, L. R., Feldstein, H. S., Haddad, L., and Peña, C. (1995). Women: The key to food security (Food Policy Report). Washington, DC: International Food Policy Research Institute. doi: 10.1177/156482659601700116

CrossRef Full Text | Google Scholar

Quisumbing, A. R., Sproule, K., Martinez, E. M., and Malapit, H. (2021). Do tradeoffs among dimensions of women's empowerment and nutrition outcomes exist? Evidence from six countries in Africa and Asia. Food Policy 100, 102001. doi: 10.1016/j.foodpol.2020.102001

PubMed Abstract | CrossRef Full Text | Google Scholar

Rahman, A. (2018). Cultural norms and son preference in intrahousehold food distribution: a case study of two Asian rural economies. Rev. Income Wealth 65, 415–461. doi: 10.1111/roiw.12356

CrossRef Full Text | Google Scholar

Reinbott, A., Schelling, A., Kuchenbecker, J., Jeremias, T., Russell, I., Kevanna, O., et al. (2016). Nutrition education linked to agricultural interventions improved child dietary diversity in rural Cambodia. Br. J. Nutr. 116, 1457–1468. doi: 10.1017/S0007114516003433

PubMed Abstract | CrossRef Full Text | Google Scholar

Renkow, M., Hallstrom, D. G., and Karanja, D. D. (2004). Rural infrastructure, transactions costs and market participation in Kenya. J. Dev. Econ. 73, 349–367. doi: 10.1016/j.jdeveco.2003.02.003

CrossRef Full Text | Google Scholar

Rheinlander, T., Samuelsen, H., Dalsgaard, A., and Konradsen, F. (2010). Hygiene and sanitation among ethnic minorities in Northern Vietnam: does government promotion match community priorities? Soc. Sci. Med. 71, 994–1001. doi: 10.1016/j.socscimed.2010.06.014

PubMed Abstract | CrossRef Full Text | Google Scholar

Ruel, M. T., Alderman, H., and Maternal and Child Nutrition Study Group (2013). Nutrition-sensitive interventions and programmes: How can they help to accelerate progress in improving maternal and child nutrition? Lancet 382, 536–551. doi: 10.1016/S0140-6736(13)60843-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Ruel, M. T., Levin, C. E., Armar-Klemesu, M., Maxwell, D., and Morris, S. S. (1999). Good care practices can mitigate the negative effects of poverty and low maternal schooling on children's nutritional status: Evidence from Accra. World Dev. 27, 1993–2009. doi: 10.1016/S0305-750X(99)00097-2

CrossRef Full Text | Google Scholar

Senauer, B., Garcia, M., and Jacinto, E. (1988). Determinants of the intrahousehold allocation of food in the rural Philippines. Am. J. Agric. Econ. 70, 170–180. doi: 10.2307/1241987

CrossRef Full Text | Google Scholar

Sharma, I. K., Di Prima, S., Essink, D., and Broerse, J. E. W. (2021). Nutrition-sensitive agriculture: a systematic review of impact pathways to nutrition outcomes. Adv. Nutr. 12, 251–275. doi: 10.1093/advances/nmaa103

PubMed Abstract | CrossRef Full Text | Google Scholar

Shively, G., and Sununtnasuk, C. (2015). Agricultural diversity and child stunting in Nepal. J. Dev. Stud. 51, 1078–1096. doi: 10.1080/00220388.2015.1018900

CrossRef Full Text | Google Scholar

Sibhatu, K. T., and Qaim, M. (2017). Rural food security, subsistence agriculture, and seasonality. PLoS ONE 12, e0186406. doi: 10.1371/journal.pone.0186406

PubMed Abstract | CrossRef Full Text | Google Scholar

Socialist Republic of Vietnam (2012). National Nutrition Strategy for 2011-2020, With a Vision Toward 2030. Hanoi: Medical Publishing House.

StataCorp (2017). Stata: Release 15. Statistical Software. College Station, TX: StataCorp LLC.

Stifel, D., and Minten, B. (2017). Market access, well-being, and nutrition: evidence from Ethiopia. World Dev. 90, 229–241. doi: 10.1016/j.worlddev.2016.09.009

PubMed Abstract | CrossRef Full Text | Google Scholar

Svedberg, P. (1990). Undernutrition in Sub-Saharan Africa: is there a gender bias? J. Dev. Stud. 26, 469–486. doi: 10.1080/00220389008422165

CrossRef Full Text | Google Scholar

Taren, D., and Alaofe, H. (2013). Feed the Future Learning Agenda Literature Review: Improved Nutrition and Diet Quality. Rockville, MD: Westat.

Teerawichitchainan, B., and Phillips, J. F. (2008). Ethnic differentials in parental health seeking for childhood illness in Vietnam. Soc. Sci. Med. 66, 1118–1130. doi: 10.1016/j.socscimed.2007.10.020

PubMed Abstract | CrossRef Full Text | Google Scholar

Thurstans, S., Opondo, C., Seal, A., Wells, J., Khara, T., Dolan, C., et al. (2020). Boys are more likely to be undernourished than girls: a systematic review and meta-analysis of sex differences in undernutrition. BMJ Glob. Health 5, e004030. doi: 10.1136/bmjgh-2020-004030

PubMed Abstract | CrossRef Full Text | Google Scholar

Tinh, V. X. (2002). Looking for Food: The Difficult Journey of the Hmong in Vietnam (anthropological perspectives on food security), Land Tenure Center Brown Bag Seminar Paper, University of Wisconsin-Madison. Available online at: https://nelson.wisc.edu/ltc/publications/vxt0211loo.php (accessed July 25, 2018).

Tran, D. V. (2003). Culture, environment, and farming systems in Vietnam's Northern Mountain Region. Southeast Asian Stud. 41, 180–205. Retrieved from: https://kyoto-seas.org/pdf/41/2/410203.pdf (accessed October 30, 2018).

Turner, R., Hawkes, C., Jeff, W., Ferguson, E., Haseen, F., Homans, H., et al. (2013). Agriculture for improved nutrition: The current research landscape. Food Nutr. Bull. 34, 369–377. doi: 10.1177/156482651303400401

PubMed Abstract | CrossRef Full Text | Google Scholar

United Nations Children's Fund (UNICEF) (2013). Improving Child Nutrition: The Achievable Imperative for Global Progress. New York, NY: United Nations Children's Fund (UNICEF).

United Nations Population Fund (UNFPA) and United Nations Children's Fund (UNICEF) (2018). Understanding Child Marriage in Viet Nam. Hanoi: United Nations Population Fund (UNFPA), and United Nations Children's Fund (UNICEF).

Variyam, J. N., Blaylock, J. R., Smallwood, D. M., Food and Rural Economics, Division, and Basiotis, P. P. (1998). USDA's Healthy Eating Index and nutrition information (Technical Bulletin 33588). Washington, DC: Center for Nutrition Policy and Promotion, USDA.

Walker, S. P., Chang, S. M., Wright, A., Osmond, C., and Grantham-Mcgregor, S. M. (2015). Early childhood stunting is associated with lower developmental levels in the subsequent generation of children. J. Nutr. 145, 823–828. doi: 10.3945/jn.114.200261

PubMed Abstract | CrossRef Full Text | Google Scholar

Wamani, H., Astrom, A. N., Peterson, S., Tumwine, J. K., and Tylleskar, T. (2007). Boys are more stunted than girls in sub-Saharan Africa: a meta-analysis of 16 demographic and health surveys. BMC Pediatr. 7, 17. doi: 10.1186/1471-2431-7-17

PubMed Abstract | CrossRef Full Text | Google Scholar

Waswa, L. M., Jordan, I., Herrmann, J., Krawinkel, M. B., and Keding, G. B. (2015). Community-based educational intervention improved the diversity of complementary diets in western Kenya: Results from a randomized controlled trial. Public Health Nutr. 18, 3406–3419. doi: 10.1017/S1368980015000920

PubMed Abstract | CrossRef Full Text | Google Scholar

WHO Expert Consultation (2004). Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 363, 157–163. doi: 10.1016/S0140-6736(03)15268-3

PubMed Abstract | CrossRef Full Text | Google Scholar

Wibowo, Y., Sutrisna, B., Hardinsyah, H., Djuwita, R., Korib, M. M., Syafiq, A., et al. (2015). Relationship between intra-household food distribution and coexistence of dual forms of malnutrition. Nutr. Res. Pract. 9, 174–179. doi: 10.4162/nrp.2015.9.2.174

PubMed Abstract | CrossRef Full Text | Google Scholar

Wijk, S., and Everaarts, A. P. (2007). The Market for Vegetables in North Vietnam. Wageningen: Wageningen University.

World Bank (2019). Drivers of Socio-Economic Development Among Ethnic Minority Communities in Vietnam. Hanoi: World Bank. doi: 10.1596/32307

CrossRef Full Text | Google Scholar

World Health Organization (WHO) (2010a). Nutrition Landscape Information System (NLIS) Country Profile Indicators: Interpretation Guide. Geneva: World Health Organization (WHO).

World Health Organization (WHO) (2010b). WHO Anthro for Personal Computers Manual: Software for Assessing Growth and Development of the World's Children. Geneva: WHO.

World Health Organization and Food and Agriculture Organization of the United Nations (2005). Fruit and Vegetables for Health: Report of a Joint FAO/WHO Workshop, 1-3 September, 2004, Kobe, Japan. Geneva: World Health Organization and Food and Agriculture Organization of the United Nations.

World Health Organization and United Nations Children's Fund (2006). Core Questions on Drinking-Water and Sanitation for Household Surveys. Geneva: World Health Organization and United Nations Children's Fund.

Wu, H., Ma, C., Yang, L., and Xi, B. (2021). Association of parental height with offspring stunting in 14 low- and middle-income countries. Front. Nutr. 8, 650976. doi: 10.3389/fnut.2021.650976

PubMed Abstract | CrossRef Full Text | Google Scholar

Ye, J., and Pan, L. (2016). Concepts and Realities of Family Farming in Asia and the Pacific (Working Paper No. 139). Brasilia: International Policy Center for Inclusive Growth.