Household Food Insecurity is Associated with Children�s
Nutritional Status: A 2015 Cross-Sectional Study of
Malawi

Ahmadi D

Special Article - Children’s Nutrition

Austin J Nutri Food Sci. 2019; 7(7): 1131.

Household Food Insecurity is Associated with Children’s Nutritional Status: A 2015 Cross-Sectional Study of Malawi

Ahmadi D*

McGill Institute for Global Food Security, Canada

*Corresponding author: Davod Ahmadi, Research Assistant, McGill Institute for Global Food Security, Quebec, Canada

Received: October 09, 2019; Accepted: November 07, 2019; Published: November 14, 2019

Abstract

Child malnutrition is a serious public health problem in sub-Saharan Africa and presents various challenges for communities with a high rate of child mortality. This is especially a concern in Malawi, where the rates of child stunting are higher than most other areas in the world. The main purpose of this study is to investigate the association between household food security and child anthropometric indicators in Malawi. Data were obtained from “Feed the Future Malawi Interim Survey in the Zone of Influence”. “The Feed the Future ZOI in Malawi covered at baseline and in the first interim assessment includes the rural areas of seven ZOI districts in the Central and Southern regions of Malawi: Balaka, Dedza, Lilongwe, Machinga, Mangochi, Mchinji, and Ntcheu”. Different statistical analyses, such as descriptive and regression, formed basis of this study. Linear regression analysis was performed to determine the association between maternal food security and the outcomes variables of child underweight, stunting and wasting measured using z-scores. This study showed that household greater food security was associated with a lower risk underweight, but not with other indicators. In our adjusted model, child’s sex, age and household highest education category were significantly associated with all anthropometric indicators. This research is important to inform food policy in Malawi for the improvement of child nutrition with a focus on food security initiatives.

Keywords: Food Security; Child growth; Dietary diversity; Malawi

Introduction

Child under nutrition is a major public health problem, particularly in developing countries [1]. Approximately one-third of all undernourished children globally live in sub-Saharan Africa (SSA) [2]; and chronic malnutrition has been a persistent problem for young children in the region [3]. In developing countries, the mortality rate of undernourished children is much higher than that of their wellnourished counterparts [4], as evidenced by reported associations between child malnutrition and child death in the literature [5].

Four principal anthropometric indicators used to measure malnutrition are stunting, wasting, underweight and BMI-for-age. Stunting refers to a condition in which a child has insufficient height for their age (H/A), while wasting describes a situation in which a child has failed to achieve sufficient weight for their height (W/H). Underweight is defined as a state in which a child’s weight is less than expected for their age (W/A) [6]. Lastly, BMI-for-age is defined as children’s weight in kilograms divided by the square of height in meters [7].

Child under nutrition remains a significant public health problem in Malawi [8] where it is amongst the countries with the highest prevalence of stunted children in SSA [9]. However, Malawi has made notable improvements in child health, including the reduction of infant and under-five mortality. Notably, the prevalence of stunting and underweight has been decreased since 1992; with the remarkable decrease in stunting in 2010 (47%). However, changes in the prevalence of wasting and overweight were small over this same time period [10].

Child nutrition is rooted in a number of factors such as households as well as communities characteristics[11]. These factors such as national and household food security, heavy maternal workload, maternal dietary diversity, HIV, children’s illness, child’s age are associated with child malnutrition in Malawi [12,13].

Related Literature

Malawi ranks 73rd out of 104 countries on the Global Hunger Index [14]. According to a study by Department of Disaster Management Affairs (DoDMA), Government of Malawi, and the United Nations Office in Malawi (2017), 6.5 million people, about 39% of the total population was at risk of food insecurity (FIS) in 24 of the 28 districts. However, the risk of FIS reached 6.7 million people in 2016 [15].

In a study, Sassi (2013) argued that “maize production and prices are at the center of the Malawi’s food security policy, which in the recent years has contributed to accentuating supply shortages of maize on the domestic market and price increases” (P: 5). Notably, major changes in the price of maize throughout the year affect household food security. Food price is based on seasonal patterns which is dependent on rain-fed agriculture [16]. Thus, household food security and child nutrition are most vulnerable to these fluctuations during the lean season (the dry, summer season September to February) [17].

Among factors influencing children’s malnutrition, there are some factors such as child’s age, gender, parental education, family assets, sanitation, and health service availability which play more important roles [11]. A study by De Groot et al., (2017) indicated that children’s characteristics are important in explaining nutritional outcomes of children in Malawi [18]. Another study focusing on Malawi by Chikhungu et al., (2014) indicated that underweighting is increased among children with illnesses (i.e., diarrhea, lower respiratory infections, fever, stomachache and upper respiratory infection) compared to children that were not ill in Malawi [19].

In terms of maternal education, a study in Malawi has shown that stunting is significantly associated with mothers’ education. Child stunting was lowest among children whose mothers had senior secondary education and above [20]. Also, another study by Arimond and Ruel (2004) indicated that the association between dietary diversity and child nutritional status might vary depending on maternal characteristics (height, BMI, education and number of prenatal care visits) [21].

Women’s workloads and energy expenditure further increase the risk of child malnutrition. Women’s workloads also made breastfeeding and responsive feeding difficult [22]. In fact, most of women must fetch water after giving birth, which can affect the quantity and quality of childcare practices [23].

Beside Malawian women, men’s attendance in child’s health has implications for child’s nutrition. For instance, when men attend in health clinics, they are provided with information that encourage them to appreciate the importance of their participation in child nutrition [24].

Sanitation plays the important role in child nutrition. For instance, a study by Ruel-Bergeron et al., (2015) argued that better wash practices contribute to reduction in stunting in Malawi [25].

HIV epidemic still has an impact on child mortality in Malawi [26]. A study by Chihana et al., 2015 indicated that the prevalence of HIV was between 10-14% in Malawi (between 2006 to 2011) and around one million adults live in Malawi [27] and their children are estimated to be infected. Specifically, HIV prevalence among women in the reproductive age group was around 13% over the same time period, and 30,000 newborns were infected through mother-to-child transmission every year. In fact, child mortality is affected indirectly through maternal death or illness as a result of HIV/AIDS [28].

The main objective of this study was to explore the association between household food security status (using Household Hunger Scale (HHS)) and children’s anthropometric indicators. The significance of this study was to understand how the determinants associate with all four indicators when examined together.

A number of covariates such as households’ characteristic (i.e., water type, sanitation type and HH highest education category), women’s dietary diversity, women’s workloads, child’s sex, and child’s age were used to assess the associations with the aforementioned anthropometric indicators in Malawi.

Methods

Data Sources

Data were obtained from “Feed the Future Malawi Interim Survey in the Zone of Influence” [1]. “The Feed the Future ZOI in Malawi covered at baseline and in the first interim assessment includes the rural areas of seven ZOI districts in the Central and Southern regions of Malawi: Balaka, Dedza, Lilongwe, Machinga, Mangochi, Mchinji, and Ntcheu” [29]. The objective of the Malawi Feed the Future program is to sustainably reduce poverty and hunger and to improve the nutrition of women and children [1].

Sample population

“The sample for the Malawi ZOI interim assessment comprises rural areas of seven districts distributed across Central and Southern regions: Balaka, Dedza, Lilongwe, Machinga, Mangochi, Mchinji, and Ntcheu. A total of 861 households in 43 clusters were interviewed by the National Statistical Office of Malawi (NSO) in the rural areas of Dedza, Lilongwe, Mangochi, Mchinji, and Ntcheu. A total of 210 households were interviewed by ICF International in the rural areas of Balaka and Machinga (See appendix for detail of sample calculation)” (Figure 1) [29].

Figure 1: Map of Malawi: Feed the Future 2015, Interim Assessment Coverage area.

    
    
    Figure 1: Map of Malawi: Feed the Future 2015, Interim Assessment
Coverage area.

“The questionnaire used for the ZOI interim survey in Malawi was based on the population-based survey instrument for feed the Future ZOI indicators for the interim assessments. The questionnaire was translated into two native languages spoken by 10 percent or more of the population in the ZOI. In Malawi, the questionnaire was translated into Chichewa and Yao. The quality of the translations was assured by using a team translation approach with back translation from the main translation. Translations were incorporated into the data entry program on the tablet computers that were used for data collection in the households” [29].

Outcome variables

Inclusion criteria for anthropometric measurement of children were for Children under five years of age as well as children who permanently live with family members in households. Key variables to be measured in the children’s anthropometric survey included sex, age, height/length and weight. Anthropometric measurements (length/height and weight) of children (5‹) months of age were taken to calculate their nutritional status.

The outcome variables (as continuous z-scores) were based on children’s anthropometric measurements, including underweight (weight-for-age ‹ -2 SD), stunting (height-for-age ‹-2 SD), and wasting (weight-for-height ‹ -2 SD) (Table 1) [30].

Table 1: Test of data normality.




  
     
    Kolmogorov-Smirnov
    Shaprio-Wilk
  
  
     
    Statistics
    df
    sig
    Statistics
    df
    sig
  
  
    Child    underweight (weight for age) Z score
    0.022
    821
    2.00
    0.996
    821
    0.027
  
  
    Child    stunting (height for age) Z score
    0.310
    821
    0.061
    0.989
    821
    0.000
  
  
    Child    wasting (weight for height) Z score
    0.034
    821
    0.029
    0.986
    821
    0.000
  
  
    Reference:    Micro data analysis from feed the Future Malawi Interim Survey in the zone of    influence, 2015.



Table 1: Test of data normality.

Exposure variables

The Household Hunger Scale (HHS) was used to measure households’ food security status (Table 2). This index contains six items focusing on the access dimension of food security. “These questions were designed to represent varying levels of food insecurity while also reflecting three domains perceived as central to the experience of food insecurity cross-culturally: 1) anxiety about household food supply; 2) insufficient quality, which includes variety, preferences, and social acceptability; and 3) insufficient food supply and intake and the physical consequences. The frequency-ofoccurrence questions ask how often a reported condition occurred during the previous 4 weeks (30 days): rarely, sometimes, or often” (Table 2). The answers were combined and based on summation. HHS was recoded as 0-1 for “food secure”, 2-3 for “moderately food insecurity (FIS)”, 4-6 for “severely food insecure” [31].

Table 2: Household Hunger Scale (HHS) characteristics.




  
    Q1
    In    the past [4 weeks/30 days], was there ever no food to eat of any kind in your    house because of lack of resources to get food?
    0 =    No
      1 =    Yes
  
  
    Q2
    How    often did this happen in the past [4 weeks/30 days]?
    1 =    Rarely (1–2 times) 2 = Sometimes (3–10 times)
      3 =    Often (more than 10 times)
  
  
    Q3
    In    the past [4 weeks/30 days], did you or any household member go to sleep at    night hungry because there was not enough food?
    0 =    No
      1 =    Yes
  
  
    Q4
    How    often did this happen in the past [4 weeks/30 days]?
    1 =    Rarely (1–2 times) 2 = Sometimes (3–10 times)
      3 =    Often (more than 10 times)
  
  
    Q5
    In    the past [4 weeks/30 days], did you or any household member go a whole day    and night without eating anything at all because there was not enough food?
    0 =    No
      1 =    Yes
  
  
    Q6
    How    often did this happen in the past [4 weeks/30 days]?
    1 =    Rarely (1–2 times) 2 = Sometimes (3–10 times)
      3 =    Often (more than 10 times)
  
  
    Reference:    Micro data analysis from feed the Future Malawi Interim Survey in the zone of    influence, 2015



Table 2: Household Hunger Scale (HHS) characteristics.

Factors such as, household water types, sanitation types, households’ highest education level, women’s age (year)(15-24, 25- 29, 30-39, and 40-49), dietary diversity, child’s sex and child’s age (months) (0-5, 6-11, 12-23, 24-59)were selected as covariates. Major limitation of this study is lacking a question for including mothers with children. However, sample includes women who mentioned they have children younger than 5 in the household they live.

Women’s dietary diversity score was used from surveys. This question is related to the consumption of starchy staples, dark green leafy vegetables, other Vitamin A-rich fruits and vegetables, other fruits and vegetables, organ meats, meat and fish, eggs, legumes, nuts and seeds, milk and milk products. Based on the summation from the above food items, a new score was produced and categorized into two groups: “0 to 3 food items” and “4 to 9 food items” (Kennedy, 2011) [30].

A question was used for women’s workload in this study. This factor is important because of the adverse impact of mothers’ work outside the home on the health and development of their infants [32]. Workload question was recorded into two categories: 0-500 minutes and greater than 500 minutes.

Analysis

Descriptive statistics were used to examine the frequency of dependent, independent and covariate variables. Four linear regression analyses were carried out to explore the association between household food security status (HHS) and their children’s anthropometric indicators with adjustment for covariates.

Results

Child, women and household (HH) characteristics

Child, maternal and households’ characteristics are presented in (Table 3). Approximately forty percent (40%) of mothers were between 15 to 24 years of age. Only ten percent (10%) of mothers had “completed secondary or more” education levels. Results for household food security status showed that fifty percent (50%) of households reported being food secure. In terms of dietary diversity, seventy-five percent (75%) of mothers ate zero to three food items. In terms of child anthropometry, around forty-two percent (26%) of children were stunted (-2SD) and sixteen percent (12%) were underweight.

Table 3: Household (HH), Women and Child (Under 5 years of age) characteristics in Malawi.




  
    HH    characteristics
    Household    Hunger Scale (HHS)
    Little    to no HH hunger (0-1)
    310    (50.5)
  
  
     
     
    Moderate    HH hunger (2-3)
    284    (46.3)
  
  
     
     
    Severe    HH hunger (4-6)
    20    (3.3)
  
  
     
    Electricity?
    Yes
    13    (2.1)
  
  
     
    Improved    Water?
    Yes
    526    (85.7)
  
  
     
    Improved    Sanitation?
    Yes
    301 (49.0)
  
  
     
    HH    Highest education category?
    Completed    secondary or more
    60    (9.8)
  
  
    Maternal    characteristics
    Maternal    age groups
    15-24
    244    (39.2)
  
  
     
     
    25-29
    129    (20.7)
  
  
     
     
    30-39
    201    (32.3)
  
  
     
     
    40-49
    48    (7.7)
  
  
     
    Education
    Completed    secondary or more
    30    (4.9)
  
  
     
    Dietary    diversity
    Meets    minimum dietary diversity (4-9)
    149    (24.1)
  
  
     
    Workload
    0-500    min
    248    (41.1)
  
  
     
    BMI
    Normal    (18.5<=BMI<25)
    451    (79.5)
  
  
    Child    characteristics
    Sex
    Girls
    404    (47.9)
  
  
     
     
    Boys
    439    (52.1)
  
  
     
    Age    (months)
    0-5
    83    (9.9)
  
  
     
     
    6-12
    92    (10.9)
  
  
     
     
    13-24
    171    (20.3)
  
  
     
     
    25-59 
    496    (58.9)
  
  
    Anthropometric    indicators
    Child    underweight (weight for age) Z score
    (-2)
    138    (16.6)
  
  
     
     
    (-3)
    22    (2.7)
  
  
     
    Child    stunting (height for age) Z score
    (-2)
    348    (42.1)
  
  
     
     
    (-3) 
    116    (14.0)
  
  
     
    Child    wasting (weight for height) Z score
    (-2)
    30    (3.7)
  
  
     
     
    (-3)
    5    (0.6)
  
  
    Reference:    Micro data analysis from feed the Future Malawi Interim Survey in the zone of    influence, 2015.



Table 3: Household (HH), Women and Child (Under 5 years of age) characteristics in Malawi.

Linear Regression analysis of the association between Household Hunger Scale (HHS) with stunting, wasting, and under weight (Table 4).

Table 4: Unadjusted linear regression analyses between anthropometric indicators and HHS and adjusting covariates (n=821).




  
     
    Underweight
      (weight    for age) 
    Stunting
      (height    for age) 
    Wasting
      (weight    for height) 
  
  
     
     
      Beta
     
      95%    CI L-H
     
      Beta
     
      95%    CI L-H
     
      Beta
     
      95%    CI L-H
  
  
    Household    Hunger Scale
    -0.087*
    -0.296;    -0.013
    -0.045
    -0.270;    0.076
    -0.072
    -0.286;    0.015
  
  
    Improved    water
    0.074
    -0.017; 0.433
    0.044
    -0.125; 0.425
    0.050
    -0.091; 0.386
  
  
    Improved    Sanitation?
    0.092*
    0.025;0.341
    0.083*
    0.008;0.392
    0.043
    -0.078;0.257
  
  
    HH    highest education category
    0.182***
    0.148;0.373
    0.126**
    0.082;0.356
    0.131***
    0.077;0.317
  
  
    Women    dietary diversity
    0.049
    -0.071;    0.300
    -0.005
    -0.242;    0.211
    0.085 *
    0.013;    0.408
  
  
    Women    Workload
    0.041
    0.000;0.001
    0.066
    0.000;0.001
    -0.010
    -0.001;0.000
  
  
    Child    Sex
    -0.054
    -0.259;    0.030
    -0.094**
    -0.430;    -0.069
    0.010
    -0.130;    0.176
  
  
    Child    age (mons)
    -0.179***
    -0.258;-0.117
    -0.279***
    -0.457; -0.283
    -0.017
    -0.096;0.058
  
  
    Reference:    Micro data analysis from feed the Future Malawi Interim Survey in the zone of    influence, 2015.



Table 4: Unadjusted linear regression analyses between anthropometric indicators and HHS and adjusting covariates (n=821).

Underweight (weight for age)

HHS was significantly associated with underweight (Beta= -0.087; 95% CI (0.296; -0.013)). However, results of the covariates showed that improved sanitation (Beta= 0.092; 95% CI (0.025; 0.341)), household highest education category (Beta= 0.182; 95% CI (0.148; -0.373)), and child’s age (Beta= -0.179; 95% CI (-0.258; -0.117)) were also significantly associated with underweight (Table 4).

Stunting (height for age)

No significant association was observed between HHS and stunting status. However, improved sanitation (Beta= 0.083; 95% CI (0.008; 0.392)), household highest education category (Beta= 0.126; 95% CI (0.082; -0.356)) child’s sex (Beta= -0.094; 95% CI (-0.430; -0.069)), and child’s age (Beta= -0.279; 95% CI (-0.457; -0.283)) were found significantly related to stunting (Table 4).

Wasting (weight for height)

Similar to the stunting, HHS was not significantly associated with wasting; however, household highest education category (Beta= 0.131; 95% CI (0.077; -0.317)) and women dietary diversity (Beta= 0.085; 95% CI (0.013; -0.408)) remained significant with wasting (Table 4).

Results for adjusted model are presented in (Table 5). Findings showed that after adjusting controlling variables, household highest education category (Beta= 0.138; 95% CI (0.076; -0.316)) and child’s age (Beta= 0.146; 95% CI (-0.260; -0.073)) were significantly associated with underweight status. Similarly, stunting status was significantly associated with household highest education category (Beta= 0.101; 95% CI (0.031; -0.319)), child’s sex (Beta= -0.104; 95% CI (-0.444; -0.059)), and child’s age (Beta= -0.226; 95% CI (-0.437; -0.208)). Furthermore, among the controlling factors, only household highest education category (Beta= 0.089; 95% CI (0.004; 0.263)) remained significant with wasting status.

Table 5: Adjusted linear regression analyses between anthropometric indicators and HHS by adjusting covariates (n=821).




  
     
    Underweight ^a.
      (weight    for age) 
    Stunting ^b.
      (height    for age) 
    Wasting ^c.
      (weight    for height) 
  
  
     
     
      Beta
     
      95%    CI L-H
     
      Beta
     
      95%    CI L-H
     
      Beta
     
      95%    CI L-H
  
  
    Household    Hunger Scale
    -0.025
    -0.192; 0.104
    -0.005
    -0.190; 0.168
    -0.028
    -0.212; 0.108
  
  
    Improved    water
    0.076
    -0.016; 0.430
    0.043
    -0.125; 0.413
    0.054
    -0.084; 0.397
  
  
    Improved    Sanitation?
    0.066
    -0.0131;0.290
    0.053
    -0.065; 0.321
    0.036
    -0.099; 0.247
  
  
    HH    highest education category
    0.138**
    0.076; 0.316
    0.101*
    0.031; 0.319
    0.089*
    0.004; 0.263
  
  
    Women    dietary diversity
    0.023
    -0.142; 0.248
    -0.014
    -0.274; 0.195
    0.058
    -0.068; 0.354
  
  
    Women    Workload
    0.016
    0.000; 0.001
    0.055
    0.000; 0.001
    -0.034
    -0.001; 0.000
  
  
    Child    Sex
    -0.044
    -0.246; 0.073
    -0.104*
    -0.444; -0.059
    0.036
    -0.099; 0.247
  
  
    Child    age (mons)
    -0.146***
    -0.260; -0.073
    -0.226***
    -0.437; -0.208
    -0.004
    -0.108; 0.098



Table 5: Adjusted linear regression analyses between anthropometric indicators and HHS by adjusting covariates (n=821).

Discussion

The main objective of this study was to investigate the association between household food security status and child anthropometric indicators. Descriptive findings indicated that prevalence of stunting (height-for-age) was high. The results of this study are in line with Demographic Health Survey, which likewise find that Malawi has one of the highest rates of stunting in Africa [33].

Findings from this study found that household food security status was significantly associated with underweighting (weight-forage). However, the strength of association was weak. Available studies have found that the severity of food insecurity affects nutrition status [34]. For instance, Humphries et al, (2015) reported lower height-forage Z-scores for children living in food insecure homes and postulated that this association was mediated by dietary diversity [35].

Child’s sex was significantly associated with stunting. Results from this study indicate that boys were more likely to be stunted compared to girls. The results align with data from the Food and Agriculture Organization [36] which demonstrate that boys are more likely to be stunted than girls in Malawi. Other studies have also found that boys are more likely to be under weight estimates compared to girls [37].

A significant relationship was found between child’s age and child’s anthropometric indicators specifically stunting and underweight. These findings concur with existing literature that has reported child’s age is an important determinants of child malnutrition [18,38].

Previous studies indicated that maternal education has a considerable impact on child health and on child mortality rates, specifically for children under two years of age [39]. A study by Makoka and Masibo (2015) indicated that stunting is significantly related to maternal education level and low education levels increase the odds of child stunting, wasting and underweight [20]. This study did not find a significant association between maternal education and children’s anthropometric indicator. However, HH highest education category was found significantly associated with all anthropometric indicators.

Women’s workload times were not significantly associated with child anthropometric indicators. However, available studies have showed that the time spent fetching water has been found to be a significant determinant of young children’s (under five years of age) health status [40]. Reduction in water fetching time affords women more time to engage in income-generating activities, which improve their household income. Thus, socioeconomic factors, and household income specifically might be associated with child’s health status.

A greater dietary diversity may be indicative of less severe food insecurity, and dietary diversity is generally considered a promising indicator of food security status [41]. However, in this study, the association between dietary diversity and wasting did achieve significance difference. Since certain foods, like maize, are more available at certain times of the year, the diet may be dominated predominately by this food crop or similar foods, thereby reducing dietary diversity, while improving food security. For instance, a report in Malawi (2017), indicated that at the beginning of lean season, food security status continues to be almost favorable at the national level. Because families consume their own productions and they have stable access to market. For instance, households can sell excess maize and use the money to buy foods at the market. However, some districts (i.e., Nsanje, Chikwawa, Mwanza, and Balaka) in Malawi experience localized dry spells and pest attacks [42].

The major limitation of this study was the use of cross-sectional data, which only provides information collected at a specific point in time and is not necessarily reflective of results over time. Moreover, the cross-sectional nature of the analysis does not allow for the estimation of casual effects, nor does it eliminate the possibility of reverse causality. Farming communities in Malawi may be more affected by access to seasonal food at different times of the year, making temporal considerations important. It is also important to note that food security and dietary diversity measures are self-reported; and therefore, results may vary from household to household. Future research should apply causal methods to investigate the impact of food security and dietary diversity on anthropometric indicators. Nevertheless, the findings of this study can inform and influence health and nutrition policy decisions at the community and national levels in Malawi by establishing maternal food security experience as a key indicator of children’s health. Further, this study lacked these variables such as income and diseases like diarrhea.

Conclusion

Results from this study revealed that household food security status in this population is associated with underweight. Moreover, this study identified several covariates, including household’s highest education category, children’s age and sex, which may be important factors to consider when estimating associations between food security and children’s nutrition status.

References

Download PDF

Citation: Ahmadi D. Household Food Insecurity is Associated with Children’s Nutritional Status: A 2015 Cross- Sectional Study of Malawi. Austin J Nutri Food Sci. 2019; 7(7): 1131.

Home

Journal Scope

Editorial Board

Instruction for Authors

Submit Your Article