Household’s Food Consumption Profile during Agricultural Mitigation Period: Burkina Faso Centre-West Region Case
Ousmane Ouedraogo; MSc*¹, Ella W. R. Compaore; PhD¹, Emile K. S.  Amouzou; PhD²,
Augustin N. Zeba; PhD³ & Mamoudou H Dicko; PhD^1
1 Department of Biochemistry-microbiology, University Joseph KI-ZERBO, Laboratory of Biochemistry, Biotechnology, Food Technology and Nutrition (LABIOTAN) 03 BP 7021 Ouagadougou 03, Ouagadougou, Burkina Faso.
² Faculty of Sciences and Technical, Laboratory of Foods Biochemistry and Nutrition, University of Kara, BP: 43 Kara, TOGO.
³ Unit of Nutrition, Institute of Health Science Research, West Regional Direction, Bobo-Dioulasso, Burkina Faso.
 

 
Introduction

Although substantial efforts have been made to prevent or handle food crisis, many countries in west Africa are still facing with food and nutritional crises due to structural or cyclical causes. In order to ensure food and nutritional security in a population, evidence-based policies and strategic investments are required in this sector. This is only possible by raising the deciders' awareness based on reliable information.
Evidence-based data on household food access mechanism or individual dietary data may be costly in term of money and time. Moreover, a high level of technical skills and capacities are needed for both data collection and analysis (Estelle and Marie, 2014). Simplified and reliable tools for individual and household diet quality assessment have been found helpful in reducing the costs related to these types of dietary surveys during the past recent years (Estelle and Marie, 2014, Food and Nutrition Technical Assistance, 2006); such as calculation of food consumption scores (FCS), food variety, and food diversity.
Food diversity, as a qualitative measure of food consumption, gives an account of the variety of foods accessible to a households, which is close to the measure of diet nutritional adequacy at an individual level (World Food Programme, 2008). Household food diversity score provides a snapshot of the economic ability of a household to access a variety of foods, as demonstrated by several studies (Gina et al., 2013, Kennedy et al., 2010).
The literature showed that the FCS was associated with caloric intake (Coates et al., 2007, Deitchler et al., 2010, Wiesmann et al., 2009). Food consumption can be used as proxy measures for the underlying nutritional status (Tiwari et al., 2013).
Another tool is the recommended energy and nutrients intake, which represents the intakes necessary for a person or a group of people to maintain good health and sustain sufficient reserves. In order to meet these needs, a food consumption national or local model is necessary by taking into account food practices and nature quality of the consumed food. Such data are not available for the Centre-West Region of Burkina Faso, where a study on household’s food and nutritional assessment was carried out. This research aimed to present the findings of this study with regard to the household dietary quality in Burkina Faso region.
Materials and Methods
Study design, location, and population: A cross-sectional study was conducted on household food consumption from February 22 to 28 in 2017. The study was conducted in the Region of the Centre-West of Burkina Faso, located at 100 kilometres from Ouagadougou, the capital, during February 2017, which is a period of agricultural mitigation (January to June) in Burkina Faso  (Conseil National de Sécurité Alimentaire, 2016). This Region includes the provinces of Boulkiemdé, Sanguié, Sissili, and Ziro. The total population of the region was estimated as1554040 in 2016 (715996 men and 838044 women) distributed among 119541 households with 87% residents in rural areas (Institut National de la Statistique et de la Démographie, 2017).
This region was the seventh poorest region of Burkina Faso with a poverty index of 41.3% in 2011 (Institut National de la Statistique et de la Démographie, 2011). In 2016, the prevalence of wasting, stunting, and underweight children was 8.8%, 25.1%, and 19.0%, respectively (Ministère de la santé, 2016).
Sampling: The number of households was estimated according to OpenEpi (version 3) proportion sample size calculation (Dean et al., 2013). The hypothesis was that at least 50% of the households will have poor FCSs.
The inclusion criterion for the households was signing informed consent forms.  The exclusion criteria included having no consent to participate in the study and having no ability to answer the questions. In each household, the person in charge of food preparation was selected for the study on behalf of the household.
Ethical considerations: The study was approved by the Ethics Committee for Health Research of Burkina Faso. The study objectives were clearly explained to the participants, selected household heads, and local authorities.
Data collection process and instruments: Investigators and previously trained supervisors collected data from households. The face-to-face interview with the concerned people was used in households using pre-tested questionnaires. In each selected household, an individual interview was conducted with the person in charge of food preparation in the household. Socio-demographic information, economic data, household food consumption, and food sources were collected. A qualitative 7-day food consumption recall was used to determine the number of days each food items or groups were eaten within the household during the previous 7 days. The number of meals per day was also collected. The 7-day recall seems to be the most appropriate recall period to capture information about the household’s habitual diet. A recall period longer than 7 days has proved to be problematic due to remembering difficulties. A shorter recall period would risk missing foods served habitually but infrequently at the household level.
The weekly special days (market, feasts, or celebrations) and normal days were both included in the recall. Yet, long periods of special diet days like Ramadan, other fasting periods, or special long festivities were excluded from the recall. Food items consumed in very small quantities (15 grams or less of fish or milk powders, oilseeds, nuts, oils/fats/butter, and 3 tiles or less of sugar in beverages, etc.) were referred to as condiments, recorded separately, and not included in the FCS. The food sources retained for survey were purchase, own production, fishing/hunting, goods/services trade, barter, borrowing, gift, collection/picking, food aid, etc.
According to the African and Burkina Faso food composition tables, a list of 23 food items/groups were surveyed (Barbara et al., 2012, Ministère de la santé, 2007). The food items were aggregated in eight food groups, including cereals, tubers and roots; legumes and nuts; meat, fish, poultry, and eggs; dark green leafy vegetables; fruit; oils and fats; milk and dairy products; and sugar/sugar products. Others information included questions related to household characteristics (household head gender, age, marital status, household income resources, drink water sources, etc.).
Study variables: Socio-economic index was a group of the electro-household goods, fuel used for kitchen, household equipment, animals possession, land possession, and household water sources. This index was transformed in 3 balanced effective classes in order to define some household with low, medium, and high socioeconomic levels. Meal frequency is a crude indicator of food consumption because households might adjust the quantity and quality of their foods. Sources of food consumed are collected to obtain a more comprehensive understanding of the household food availability and access.
Based on the information about the food consumption during the previous 7 days, the number of food groups consumed in the household were established and used for the household. The consumption days of each food group were summed and the maximum number was seven.
The FCS is a composite indicator of dietary diversity, food consumption frequency, as well as nutrient intake by a household. It is intended to capture both diet quantity and quality.
The household food consumption score was calculated by multiplying each food groups' weight (ranging from 0.5 to 4) by frequency (ranging from 0 to 7). The highest weight was attributed to foods with relatively high energy, good quality protein, and a wide range of micro-nutrients that can be easily absorbed.
The score values calculated for each household were reported on a scale ranging from 0 to 112. The standard thresholds were used to determine three classes of households’ food consumption (poor, borderline, and acceptable).
The poor consumption class (severe food insecurity) was related to households whose FCS was less than or equal to 21. The borderline consumption class (moderate food insecurity) included households whose FCS was higher than 21 and less than or equal to 35.
The acceptable consumption class (food safety) composed of households with FCS higher than 35. The consumption frequency of food groups rich in specific nutrients (vitamin A, iron, and protein) was determined by adding food consumption frequency from subgroups rich in nutrients.
To collect more information on consumption rate of households in specific nutrient-rich food groups (iron, vitamin A, protein), an analysis of the nutritional quality of the food consumption was made according to the WFP methodology (World food programme, 2015) .
Vitamin A-rich foods are: dairy, offal, eggs, vitamin A-rich vegetables, green leafy vegetables, and vitamin A-rich fruits. Protein-rich foods are: legumes, dairy products, lean meat, offal, fish, and eggs. Bioavailable iron-rich foods are: lean meat, offal, and fish.
In this study, we assessed the association of FCS and consumption frequency of the specific nutrient rich food groups with socio-demographic and economic variables.
Ethics and informed consent: Study protocol was validated by a committee of doctoral school of sciences and technologies in the University of Ouaga 1 Pr Joseph Ki-Zerbo. Health authorities in the Region of Centre-West were given authorization prior to the data collection. Informed consent forms were obtained before any inclusion of a household or an individual in the study. The protocol of this study was approved by Burkina Faso ethics committee of health research.
Treatment and data analyses of data: Data were analysed with IBM-SPSS version 20.0 (IBM Corp., 2011). Socio-economic index was calculated by principal component analysis. The electro-household goods, fuel used for kitchen, household equipment, animals’ possession, land possession, and household water sources were used in determination of the socioeconomic status index. Socioeconomic index was the first component obtained after extraction and correlation of the matrix method. This index was transformed in 3 balanced effective classes in order to define some households with low, medium, and high socioeconomic levels. Later, bivariate descriptive analyse was used. Variables were expressed as frequencies, percentages, and mean ± standard deviation (SD) with one decimal and confidence intervals (95%). Differences were considered significant at P-value < 0.05 for all tests.
Results
Socio-demographic characteristics of the surveyed households: The sample consisted of 985 households identified in 37 villages and 3 towns in the Centre-West Region.
As shown in Table 1, 85% of the households were headed by men, 66.5% of the household heads were uneducated, and only 12.7% of them finished the primary school. Most of the household heads (70.5%) were monogamous, while 23.4% of them were polygamous.
Half of the households (50%) had three income sources, while the others (42% and 8%) had two and one income sources, respectively. Drinking water source was fountain in 53% of the households, while 9% were still drinking water from unprotected duged wells (Table 1).
Hoseholds’ food consumption characteristics
In urban areas, the number of meals consumed per day was three or more in 74% of the households, while this rate was about 50% in adults in rural area .
In the last seven days, all households (100%) consumed grain food groups and its derivatives including tubers and legumes, 94% of households consumed meats, fish, poultry, and eggs, 90% of households consumed vegetables, 83% of households consumed legumes, 73% of households consumed sugars and sugar products, 65% of households consumed oils and fats, 35% of households consumed fruits, and 30% of households consumed milk and dairy products (Figure 1).
Regarding the food frequency consumption, analysis indicates that, seven days before the survey, households consumed mainly cereals' group (51.3%), (Figure 2).
Food consumed sources
According to Table 2, corn and millet/sorghum came from the own-production and purchases' groups. Yet, rice, fonio (Digitaria exilis), roots/tubers, poultry, meat, eggs, and legumes came from the purchases and own-production group. Majority of the food groups containing bread, wheat, fish, milk, and dairy products were provided by purchase. Sugar, sugar products, and pasta came from purchase in households that had consumed them. Vegetables/leaves, fees/dried fruits came from purchase, and collection/piking came from own production and other sources.
Food consumption score: The FCS mean±SD and median were 38.1 ± 23.2 and 31.5 (Table 3), respectively.
As presented in Table 3, the FCS was poor (FCS ≤ 21) in 28% of the households, at borderline (21 < FCS ≤ 35) in 28% of the households, and at acceptable level (FCS > 35) in 44% of the households. Almost 56% of households had limited access to food during the agricultural mitigation period and food insecurity.
According to Table 4, vitamin A-rich and protein-rich food groups' consumption was more frequent in Sanguié than other provinces (P < 0.001). Consumption of iron-rich food groups was more frequent in Ziro households than other provinces (P < 0.001). However, urban households consumed vitamin A-rich and protein-rich food groups more frequently than the rural areas (P < 0.001; P = 0.032). No significant difference was observed between rural and urban areas concerning the consumption frequency of iron- rich food groups (P = 0.215). A significant difference was seen among provinces (P < 0.001). However, increased consumption frequency of nutrients-rich food groups (vitamin A, iron, and protein) increased FCS and vice versa.
FCS, socio-demographic, and economic characteristics:
According to Table 5, people living in households headed by men had a higher food consumption rate (45%) than those headed by women (42%). The people most vulnerable to food insecurity were those living in households having one (96%) or two (79%) sources of income, respectively. More than 40% of households had at least 3 sources of income and were at the borderline to acceptable levels with regard to food consumption score. The possession of animals was decisive in food insecurity; households with animals were less affected by food insecurity than those who had none.
According to our results, 88% of the households who have no animals, were food insecure. Among households who practiced market gardening, 66% had an acceptable level of food consumption, while 37% of household have never practiced gardening (P < 0.001). The education level (P = 0.10) of the household head did not have a significant effect on the households' food situation (P = 0.06).  In rural areas, the proportion of households in the situation of severe food insecurity was 28% against 26% in urban areas (P = 0.29).