Energy and nutrient contribution of different food groups to the dietary intake of 6- to <9-month-old infants in a low socioeconomic community in North West Province, South Africa

Objective: A study was undertaken to determine the energy and nutrient contribution of different food groups to the dietary intake of 6- to <9-month-old infants. Design: An observational study was conducted using baseline data of a preliminary randomised controlled trial that aimed to determine the effect of egg consumption on infant growth. Setting and subjects: Participants resided in a peri-urban community (Jouberton) in North West province, South Africa. The study included 6- to <9-month-old infants (n = 155); 24-hour dietary recall data were available for n = 144. Results: Most infants consumed either two (29.2%) or three (42.4%) out of eight food groups. The grains/roots/tubers group was consumed by 95.8% of infants; for consumers thereof, it contributed 75.5% of iron, 53.0% of thiamine and 42.5% of folate. Breast milk and dairy were consumed respectively by 64.4% of infants. For breastfed infants, breast milk was the major contributor of energy and fat, and some micronutrients (calcium, zinc, vitamin A, vitamin C, niacin and riboflavin); but they had lower intakes (p < 0.05) for all micronutrients except vitamin A compared with non-breastfed infants. For consumers (16.7%) of animal-source foods (ASFs), these contributed 42.8% for vitamin B12 and 33.4% for protein; and intake of protein, riboflavin and vitamin B12 was higher (p < 0.05) for consumers compared with non-consumers. The least consumed food groups were legumes (0.7%), flesh foods (6.9%) and eggs (10.4%). Conclusion: Grains/roots/tubers, dairy and breast milk made a major contribution to the intake of key nutrients. Animal-source foods were not consumed frequently, but for consumers thereof made a substantial contribution as well. Recommendation: Strategies to improve dietary diversity should encourage continued breastfeeding, aim to increase intake of food groups not frequently consumed and promote locally available food.


Introduction
Good nutrition and optimal feeding practices during infancy ensure child growth and development.The complementary feeding period is the largest part of the 'first 1000 days of life', which is considered a window of opportunity for preventing malnutrition and its consequences. 1At age six months, nutrient-rich complementary foods should be introduced, along with continued breastfeeding, to meet infants' high nutrient requirements. 2However, complementary foods in low-and middle-income countries (LMICs) are often nutritionally inadequate and are deficient in several micronutrients, with iron, zinc and calcium being considered 'problem nutrients'. 3,4oor diet and inappropriate complementary feeding practices are the most common drivers of malnutrition in early childhood. 5Furthermore, growth faltering commonly occurs between the ages of 6 and 24 months in LMICs. 6Hence, promoting optimal nutrition during the complementary feeding period is a health priority. 7etary diversity has been recognised as an important indicator of a healthy diet. 8,9Consuming a diversified diet that includes a variety of food items increases the likelihood of nutrient adequacy among infants and young children. 10According to the most recent guidelines for measuring minimum dietary diversity (MDD), 6-to 24-month-old children should consume at least five out of eight food groups. 11Several studies have shown that MDD is often not achieved in LMICs, 8,[12][13][14][15] and complementary foods are mostly 'grains, roots and tubers', followed by a few animal products and a few fruits and vegetables. 8,14,15aching MDD is a challenge in South Africa. 13The 2016 South African Demographic and Health Survey (SADHS) showed that only 23% of 6-to 8-month-old infants met the criteria for MDD.The 2016 SADHS further showed that only 23% of 6-to 23-month-old children received a minimum acceptable diet (which considers breastfeeding status, numbers of meals and dietary diversity), with the proportion of children receiving a minimum acceptable diet being lowest in the lower wealth quintiles. 16 South Africa, fortified infant cereal and maize meal are frequently consumed during infancy. 17While some information is available on the nutrient contribution of fortified staple foods and commercially available infant products to dietary intake in infants and young children, 18 information on the potential nutrient contribution of the other food groups has not been widely investigated.The aim of this study was to determine the contribution of different food groups to the total dietary intake of 6-to <9-month-old infants in a low socioeconomic community in North West province, South Africa.

Study population and study design
This study was conducted in Jouberton, which is a peri-urban area in the greater Matlosana Klerksdorp municipality in North West province, South Africa.The study area is 200 km from Johannesburg.This paper is part of a preliminary study for the Eggcel-growth study, which aims to determine the effect of providing one egg per day for 6 months on infants' growth, motor development, micronutrient and morbidity status.The required sample size for the Eggcel-growth study is 500 participants, and enrolment started in February 2020.However, in the last week of March 2020, enrolment of participants was stopped because South Africa's national lockdown regulations due to the COVID-19 pandemic came into effect.At the time, 155 study participants had been enrolled, and they remained in the trial.As the sample size of 155 is inadequate to measure intervention outcomes, the trial is therefore referred to as the preliminary Eggcel-growth study.
Trained fieldworkers recruited study participants (motherinfant pairs), explained the nature of the study to the mothers and performed a pre-screen to assess potential eligibility.Mothers who were interested and potentially eligible were given an information sheet to take home.Before enrolment, at age 6 to <9 months, infants were screened for final eligibility, and their mothers were asked questions to ensure that they understood what was required and comprehended the consent process.Only infants whose mother or legal guardian provided signed consent were enrolled in the study.
Exclusion criteria included known allergies/intolerances to eggs, severe anaemia (haemoglobin < 70 g/L), severe acute malnutrition (weight-for-length Z-score < -3), obvious congenital abnormalities, not born a singleton, the mother being younger than 18 years at enrolment and the mother planning to relocate within 9 months after enrolment.For this study, baseline data for the 155 participants enrolled in the preliminary Eggcel-growth study were used, with the focus on dietary intake.

Measurements
Mothers/caregivers were interviewed by trained fieldworkers in English or their home language (Setswana).Socioeconomic information (such as level of education, age, marital status and household characteristics) and early feeding practices were collected using a structured questionnaire.
Trained fieldworkers collected information on dietary intake of the infants by interviewing the mothers/caregivers using a single multi-pass 24-hour dietary recall.A standardised dietary kit was used to assist the mother/caregiver to describe and estimate the amount of food and beverages consumed by the infant during the previous day.The dietary kit contained examples of relevant food, food wrappers and containers, commonly used household utensils (e.g.spoons and bowls) and a photobook.In addition, the portion size for cooked food was estimated using the 'dish-up and measure' method.For infant cereal and formula milk, the amounts for the dry product and the liquid were recorded separately, where possible.
Food intake data recorded in household measures were converted to grams using the Food Quantities Manual for South Africa. 19The amount of breast milk intake was estimated according to age, with 675 ml for breastfed infants at age 6 months. 20Food intake data were converted to energy, macroand micronutrients using the South African food composition database 21 and STATA software (StataCorp, College Station, TX, USA).Energy and nutrient intakes for the total diet were calculated.3][24] The mean adequacy ratio (MAR) was calculated by, first, calculating nutrient adequacy ratios (NAR) for 11 micronutrients (ratio of an individual's intake to the recommended intake) and then calculating the average of the 11 NARs (with NARs higher than 1 being truncated to 1). 25 To calculate dietary diversity and determine the energy and nutrient contribution of different food groups, foods were categorised into eight food groups, namely (i) breast milk, (ii) 'grains, roots and tubers', (iii) legumes and nuts,( iv) dairy, (v) flesh Energy and nutrient contribution of different food groups to the dietary intake of 6-to <9-month-old infants foods, (vi) eggs, (vii) vitamin A-rich fruits and vegetables and (viii) other fruits and vegetables. 11For each infant, foods consumed during the 24-hour-recall period were allocated to the appropriate food group (Table 1).For each food group consumed the previous day, a score of one was given; the MDD was calculated as the number of infants who consumed at least five of the eight food groups on the day of recall.Energy and nutrient contribution for each of the respective food groups was calculated and expressed as a percentage of total intake of the nutrient.For each food group, nutrient density was calculated (amount of nutrient provided by the food group per 418 kJ/100 kcal provided by the food group).Nutrient contribution and nutrient density of food groups are reported for consumers only, where consumers are defined as those infants who consumed the specific food group during the recall period.

Statistical analysis
IBM SPSS version 27 was used for statistical analyses (IBM Corp, Armonk, NY, USA).The data were tested for normality using the Shapiro-Wilks test.As the data were not normally distributed, results are reported as the median and interquartile range (25th and 75th percentiles).To compare the differences between consumers (consumed at least one food from the food group during the recall period) and non-consumers (did not consume any food from the food group during the recall period) of different food groups, the Mann-Whitney U test was used.For each food group, Spearman correlation analysis was done between kilojoule (kJ) provided by the food group and the MAR.For all analyses, the significance was set at p < 0.05.

Ethical consideration
This study was approved by the Human Research Ethics Committee of North-West University (NWU-00258-21-A1).For the study to be done in the community, approval was sought through community-based structures.Mothers/caregivers gave informed consent before any study processes started.

Sociodemographic and household characteristics
A total of 155 participants were enrolled in the study (Table 2).The median age of the infants was 7 months; 50.3% of infants were males and 49.7% females.The interviewees were mostly the mother of the infant (93%), only (9%) were married and most (80.6%) had an education level of grade 10 or higher.The median age of the mothers/caregivers was 28 and the median number of people in the household was 6.Most households had access to electricity (86.4%) and to tap water either inside (34.6%) or outside (59.5%) their home.
In total, 43% of the infants had been introduced to solid foods at age 3 months or younger.Commercial infant cereal was the most commonly given first solid food (> 50% of infants).

Total dietary intake
From the 155 infants enrolled in the study, dietary intake was not available for 11 infants as they were not in full-time care of the respondent during the full 24-hour recall period.Dietary intake data were therefore available for 144 infants.Total energy, macro-and micro-nutrient intakes are summarised in Table 3. Iron and zinc intakes were below the EAR for 58.3% and 27.1% of infants, respectively.For all the other micronutrients, except for niacin, median intake was above the AI value, and therefore the likelihood of these nutrients being deficient in the diet is assumed to be low.The median MAR of 0.9 suggest the probability of the diet being inadequate for some micronutrients as it did not reach 1.
Because of the small number of infants per food group, the legume group was excluded from all further analysis, while flesh foods and eggs were combined to create the animalsource foods (ASFs) group (16.7%; n = 24), and vitamin A-rich fruits and vegetables were combined with the other fruits and vegetables group to create the fruits and vegetables group (42.4%; n = 61).The further analysis, therefore, included only five food groups.

Percentage contribution of different food groups towards total intake of consumers
The percentage contribution of each of the five food groups towards the total energy and nutrient intake of consumers is listed in Table 4.
Breast milk was consumed during the recall period by 97 (67.4%) infants and based on an estimated intake of 675 ml Fruits and vegetables were consumed during the recall period by 61 (42.4%) and contributed 5.9% of total energy intake.For consumers thereof, the contribution of fruits and vegetables to total intake was less than 20% for all micronutrients.For those who consumed vitamin A-rich fruits and vegetables, the vitamin A contribution was 23.7%.Energy and nutrient contribution of different food groups to the dietary intake of 6-to <9-month-old infants

Total nutrient intake for consumers versus nonconsumers of different food groups
Nutrient intakes for consumers and non-consumers of different food groups are reported in Table 5.Only six infants did not consume any food from the 'grains, roots, and tubers' food group; intakes for consumers and non-consumers are therefore not reported.Consumers of breast milk had significantly (p < 0.05) higher intakes for energy and fat, but lower intakes for all micro-nutrients, except for vitamin A, compared with nonconsumers.Consumers of dairy had significantly (p < 0.05) higher intakes for fat and all micronutrients compared with non-consumers.Consumers of ASFs had significantly (p < 0.05) higher intakes for energy, protein, fat, riboflavin, niacin and vitamin B12 compared with non-consumers.Lastly, consumers of fruits and vegetables had significantly (p < 0.05) higher intakes for vitamin A, vitamin B6 and vitamin C compared with non-consumers.

Nutrient densities for consumers and non-consumers of different food groups
The median and interquartile range for nutrient densities (amount of specific nutrient per 418kJ) for consumers and non-consumers are reported in Table 6.Only six infants did not consume any food from the 'grains, roots, and tubers' food group; nutrient densities for consumers and non-consumers are therefore not reported.Infants who received breast milk had significantly (p < 0.05) lower nutrient densities protein and all micronutrients compared with infants who did not receive breast milk.Those who consumed dairy had significantly (p < 0.05) higher nutrient densities for protein and all micronutrients except vitamin A when compared with non-consumers.Consumers of ASFs had significantly (p < 0.05) higher nutrient densities for protein and vitamin B12 and significantly lower nutrient density for vitamin C compared with non-consumers.Consumers of fruits and vegetables had significantly (p < 0.05) higher nutrient densities for vitamin A and vitamin C when compared witho non-consumers.

Association of food group consumption and nutrient adequacy
The association between kilojoules provided by each food group respectively and the MAR is summarised in Table 7. MAR was positively correlated with two food groups, respectively ('grains, roots, and tubers'; and dairy; p < 0.001), suggesting that a higher intake of these two food groups is associated with higher micronutrient adequacy of the diet.

Discussion
In this paper, we describe the contribution of different food groups to dietary intake for energy, macro-and micronutrients in infants 6 to <9 months of age from a low socioeconomic community in North West province.Most infants consumed either two or three of the eight food groups on the day of recall; and only 6.3% reached the MDD.The most consumed food group was 'grains, roots, and tubers', followed by breast milk and dairy.The least consumed food groups were legumes, flesh foods and eggs, which is similar to the findings of previous studies done in different parts of South Africa. 13,14r consumers thereof, 'grains, roots, and tubers' contributed about a quarter of total energy intake and at least a third of the intake for various micronutrients.This food group includes fortified infant cereals as well as fortified maize meal, both of which are frequently consumed during infancy in South Africa. 17,26According to legislation, maize meal and wheat flour are fortified with eight micronutrients as part of the National Fortification Programme. 27Findings from a previous study done in the same area showed that for consumers thereof, maize meal contributed more than a third of total intake for iron, zinc, vitamin A, and folate at age 12 months, and infant cereals contributed more than 50% of total intake for iron and thiamine at age 6 months. 18The significant contribution of the 'grains, roots, and tubers' food group to the intake of micronutrients such as iron, thiamine, niacin, vitamin B6 and folate can therefore be attributed, at least to a large extent, to fortified maize meal and infant cereals.
The 'grains, roots, and tubers' group provided 75% of total iron intake.Although this food group was consumed by 95.8% of the study sample and included fortified maize meal/wheat flour and fortified infant cereal, more than half of the infants had an iron intake below the EAR, highlighting the difficulty of meeting iron requirements at this age.Findings from a study in Indonesia showed that consuming fortified infant foods was associated with lower dietary diversity, which limits infants' exposure to foods of differing textures and flavours, and over-reliance on fortified foods should thus be avoided. 28ring the recall period, 67.4% of the study sample received breast milk.This is very similar to breastfeeding rates reported for 6-month-old infants in North West province (71%) 18 and 9 month-old infants in the Western Cape (60%). 13It should, however, be noted that the WHO recommends continued breastfeeding up to age 2 years or beyond. 29r consumers thereof and based on an estimated intake of 675 ml, 20 breast milk was the major contributor of energy and fat intake, and contributed significantly to intakes of protein, calcium, vitamin A intake zinc, riboflavin and folate.These findings are similar to those found in Mexico 30 and China 31 for infants 6-11 months of age.However, intakes of all micronutrients, except vitamin A, were significantly lower while energy intake was significantly higher for consumers of breast milk compared with nonconsumers, resulting in lower micronutrient densities.Although it is recommended that breastfeeding should be continued up to age 2 years or beyond, it is important to introduce nutrientdense foods that will fill nutrient gaps after the age of 6 months. 2iry was consumed by 67.4% of infants during the recall period.For those who consumed dairy (including formula milk), this contributed to nearly half of calcium and riboflavin intake; and, compared with non-consumers, the nutrient densities were higher for all micronutrients except for vitamin A. Inclusion of formula milk in this food group probably contributed to the higher nutrient density of micronutrients for consumers thereof.Furthermore, results of the study showed that higher consumption of dairy is associated with higher micronutrient adequacy of the diet.The substantial contribution of dairy to nutrient intake was also reported for a study in Ethiopia, 32 which showed dairy to be the main contributor of protein, vitamin A, vitamin C and calcium.
Although dairy contributed nearly half of calcium intake, the median calcium intake for the total sample was substantially lower than the AI.In a previous study in the same area, 75% of 12-to 18-month-old children had low intakes of calcium, 18 while nutrient density of the complementary diet was low for calcium for nearly all 6-to 17-month-old breastfeeding children in a study in KwaZulu-Natal. 14Low intake of calcium intake has been shown to be associated with stunting in 2-to 5-year-old children, 33 and the authors contributed this association to low intake of milk.
Fruits and vegetables were consumed during the recall period by 42.4% of the infants.For consumers thereof, the contribution to total intake was low: 16.0% for vitamin B6 and 17.3% for vitamin C. In total, 13.9% of infants consumed vitamin A-rich fruits and vegetables during the recall period, contributing 23.7% of their vitamin A intake.Despite these low contributions to nutrient intake, consumers of fruits and vegetables had significantly higher intakes for vitamin A, vitamin B6 and vitamin C, and significantly higher nutrient densities for vitamin A and vitamin C, compared with non-consumers.
The early introduction of fruits and vegetables has been shown to be correlated with a higher consumption thereof throughout childhood. 34Although the revised South African Paediatric Food-Based Dietary Guidelines recommend that from age 6-12 months children should be given dark-green leafy vegetables and orange-fleshed vegetables and fruits every day, 35 cost is a barrier for consumption of fruits and vegetables. 36It has been suggested that promoting home-produced or wild-harvested dark-green leafy vegetables, 37 such as spinach and African leafy vegetables, and indigenous plants 38 as a complementary food is a low-cost strategy to improve micronutrient intake of children.
ASFs are nutrient-dense foods and contain multiple micronutrients (iron, zinc, vitamin A and vitamin B12) that can improve the micronutrient density of the complementary diet. 39,40Only a small proportion of infants consumed ASFs.For consumers, ASFs contributed only 10% of energy intake, but nearly half of vitamin B12 and protein intake; and when compared with non-consumers, consumers had significantly higher intakes for both micronutrients.Although the contribution of ASFs to a Consumers are those infants that consumed food groups on the day of recall.Non-consumers are those infants that did not consume food groups on the day of recall.b P < 0.05 indicates significant differences and is formatted in bold type; consumers and non-consumers were compared using the Mann-Whitney U test.iron intake was relatively low (19.2%) for consumers, probably because of small amounts eaten, the bioavailability of iron is high.ASFs are a rich source of highly bioavailable hemeiron, 41 and could be beneficial in increasing the availability of iron in the diet and maintaining iron stores in breastfed infants. 42The low intake of iron in this age group is of concern, as lack of iron in the diet is one of the biggest causes of iron deficiency, which is associated with poor cognitive outcomes, infections and growth faltering. 43,44rthermore, vitamin B12 is important for brain development and cognitive function. 45ASFs, together with breast milk and dairy, are needed to meet the infant's requirements for vitamin B12, which is not found naturally in plant foods. 39A study in India showed that little or no consumption of ASFs is associated with stunting and poor linear growth in children 6-23 months of age. 46Consumption of ASF in LMICs, however, remains low, with cultural issues and affordability being the common driver for their low consumption. 37,39The most affordable nutrient-dense locally available food sources of multiple key micronutrients in South Africa include beef liver, chicken liver, small fish, milk and eggs. 37though this study showed that the consumption of the 'grains, roots, and tubers' group and the dairy group respectively were associated with higher micronutrient adequacy of total dietary intake, it is important to note that the consumption of different food groups is needed to improve the micronutrient adequacy and nutrient density of the complementary diet of children 6-23 months of age. 47The consumption of different food groups is regarded as a good predictor for dietary quality and nutrient adequacy. 10,47To achieve minimum dietary diversity, five out of the eight food groups should be consumed; 11 this was achieved for only 6.3% of the infants in the study.Identifying the most affordable food sources within each of the food groups could assist caregivers to make better food choices, 37 which may contribute to higher dietary diversity.
This study provides insight into the contribution of different food groups to nutrient intake at this young age (6-9 months), which is considered a critical time during the complementary feeding period.The study does, however, have some limitations.First, the sample size is small as recruitment had to be stopped because of COVID-19 lockdown regulations.Second, the use of a 24-hour recall does not account for day-to-day variations, 48 and quantifying breast-milk intake is challenging.Although breast-milk intake can be measured accurately using the deuterium oxide dose-to-mother technique, 49 this is not practical in a community-based research setting.Age-specific average breastmilk intake values and average nutrient content values for breast milk have been used in other studies. 50,51It is, however, acknowledged that the nutrient content of breast milk may be influenced by factors such maternal dietary intake and maternal nutritional status. 52It is therefore important that dietary intake data be interpreted within the context of the inherent limitations of dietary assessment. 48 conclusion, 'grains, roots, and tubers', dairy and breast milk were the most consumed food groups and made a substantial contribution to energy and nutrient intakes.Animal-source foods were consumed by < 20% of infants but made a substantial contribution to key nutrients for those who consumed ASF.
Strategies to improve dietary diversity should encourage continued breastfeeding, aim to increase the intake of less frequently consumed food groups and promote locally available food as complementary food.Energy and nutrient contribution of different food groups to the dietary intake of 6-to <9-month-old infants

Figure 1 :
Figure 1: Percentage of infants who consumed different food groups on the day of recall (n = 144).'Grains, roots, and tubers' include infant cereals; dairy includes formula milk; F&V: fruits and vegetables.

Table 1 :
Food-group classification for the foods and drinks reported during the 24-hour recall period

Table 3 :
Median intakes and interquartile ranges for total energy and nutrient intakes for infants at age 6 to <9 months (n = 144) DRI: dietary reference intakes; g: grams; mg: milligram; μg: microgram; IQR: interquartile range; MAR: Mean adequacy ratio; RE: retinol equivalents.a Energy intake in kilojoules of the US DRI published by the Institute of Medicine. 23age 6 months (boys 2 709; girls 2 535).b EAR of the US DRI published by the Institute of Medicine. 22c AI of the US DRI published by the Institute of Medicine. 22d AI of the US DRI published by the Institute of Medicine for calcium and vitamin D. 24

Table 5 :
Total energy and nutrient intake of consumers versus non-consumers of different food groups at age 6-9 months a

Table 6 :
Median (interquartile range) for micronutrient density (amount per 418 kJ) between consumers and non-consumers of different food groups at age 6-9 months

Table 7 :
Association of different food group consumption with the mean adequacy ratio for those who consumed the food group on the day of recall