Seasonality affects dietary diversity of school-age children in northern Ghana
Seasonality affects dietary diversity of school- age children in northern Ghana
Abdul-Razak Abizari 0 1
Fusta Azupogo 1
Miwako Nagasu 1
Noortje Creemers 1
Inge D. Brouwer 1
0 Department of Community Nutrition, School of Allied Health Sciences, University for Development Studies , Tamale , Ghana , 2 Department of Family and Consumer Sciences, Faculty of Agriculture, University for Development Studies , Tamale , Ghana , 3 Division of Human Nutrition, Wageningen University , The Netherlands
1 Editor: Leng Huat Foo, University Sains Malaysia , MALAYSIA
Background and objectives Dietary diversity score (DDS) is relatively easy to measure and is shown to be a very useful indicator of the probability of adequate micronutrient intake. Dietary diversity, however, is usually assessed during a single period and little is known about the effect of seasonality on it. This study investigates whether dietary diversity is influenced by seasonality.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
Funding: Funding for this research was received
from Nestle Foundation for Nutrition Research with
Grant number FN4042. The grant was received by
Abdul-Razak Abizari. The funders had no role in
study design, data collection and analysis, decision
to publish, or preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
Two cross-sectional surveys were conducted in two different seasonsÐdry season (Octo
ber 2010) and rainy season (May 2011) among the same school-age children (SAC) in two
rural schools in northern Ghana. The study population consisted of 228 school-age children.
A qualitative 24-hour dietary recall was conducted in both seasons. Based on 13 food groups, a score of 1 was given if a child consumed a food item belonging to a particular food group, else 0. Individual scores were aggregated into DDS for each child. Differences in mean DDS between seasons were compared using linear mixed model analysis.
The dietary pattern of the SAC was commonly plant foods with poor consumption of animal
source foods. The mean DDS was significantly higher (P < 0.001) in the rainy season
(6.95 ± 0.55) compared to the dry season (6.44 ± 0.55) after adjusting for potential
confounders such as age, sex, occupation (household head and mother) and education of
household head. The difference in mean DDS between dry and rainy seasons was mainly
due to the difference in the consumption of Vitamin A-rich fruits and vegetables between the
seasons. While vitamin A-rich fruits (64.0% vs. 0.9%; P < 0.0001) and vitamin A rich dark
green leafy vegetables (52.6% vs. 23.3%, P < .0001) were consumed more during the rainy
season than the dry season, more children consumed vitamin A-rich deep yellow, orange
and red vegetables during the dry season than during the rainy season (73.7% vs. 36.4%,
Seasonality has an effect on DDS and may affect the quality of dietary intake of SAC; in such a context, it would be useful to measure DDS in different seasons. Since DDS is a proxy indicator of micronutrient intake, the difference in DDS may reflect in seasonal differences in dietary adequacy and further studies are needed to establish this.
Dietary diversity assessed as the number of food groups consumed over a reference period has
been proposed as a potential proxy indicator for diet quality [1±3]. There is substantive
evidence from developed countries showing that dietary diversity is indeed strongly associated
with nutrient adequacy and is thus an essential element of diet quality [4±7]. Studies recently
emerging from developing countries also support the association between dietary diversity
and nutrient adequacy [8±12]. Studies from resource poor settings have also established an
association between dietary diversity and nutritional status of children [13±16] as well as
Dietary diversity is usually assessed during a single time period but may be influenced by
seasonality, especially in resource poor settings. Seasonality is known to influence the
consumption of food [20±22]. Ferguson et al [
] reported seasonal variations in the consumption
of fruits, legumes, roots, plantains and some flour among Ghanaian and Malawian preschool
children. Hudelson et al [
] also observed that adequate consumption of vitamin A rich foods
in Ghana is hindered by seasonality as green leafy vegetables are abundant only in the rainy
season. A study from Burkina Faso confirmed that seasonality has an influence on dietary
diversity score (DDS) as well as the nutritional status of women [
With seasonal fluctuation in agricultural production, populations relying heavily on
agriculture outputs with little access to animal source foods are more likely to experience seasonal
food shortages. Seasonal variations in nutrient intake have also been reported. A study in
Kenya found significant improvement in intakes of calcium, zinc, iron and folate during the
rainy season compared to the dry season among preschool children [
]. Whilst some studies
have documented significantly higher intakes of micronutrients during the post-harvest season
compared to the lean season in Burkina Faso [
], Savy et al  reported a higher DDS
with a decreased nutrient adequacy in the lean season compared to the post-harvest season in
rural Burkina Faso. This suggests there may be a lack of consistency in the pattern of seasonal
fluctuations of dietary diversity and nutrient adequacy and more studies are needed to
elucidate the effect of seasonality on DDS.
To the best of our knowledge, our present study is the first to examine the influence of
seasonality on the dietary diversity of school-age children (SAC). We hypothesized that the
dietary diversity score of SAC in Northern Ghana will vary between the dry season (October) and
the rainy season (May). The information from the present study will be useful in informing
the selection and design of context-specific strategies to reduce micronutrient deficiencies.
Materials and methods
The study was carried out in 2 primary schools in 2 rural communities in the Tolon district of
northern Ghana between October 2010 and May 2011. The 2 communities have similar
socio2 / 16
demographic characteristics. The climate of the district is relatively dry, with a single rainy
season that begins in April and ends by October. The dry season starts in October and ends in
March with maximum temperatures occurring towards the end of the dry season [
vegetation is guinea savannah characterized by tall grasses interspersed with drought resistant
trees such as the shea (Butyrospermum Parkii) and dawadawa (Parkia Biglobosa) [
Subsistence agriculture is the main stay of the people in the district and major crops cultivated
include cereals (maize, sorghum, millet and rice), legumes (soybeans, groundnuts and
cowpeas) and starchy roots and tubers (cassava, yam and potato) [
Study design and study population
This study was carried out within the framework of the TELFUN-Ghana dietary intervention
]. A sample size of 241 was determined for the trial based on another primary outcome
(iron deficiency anaemia). In brief, two communities (with primary schools) which qualified
to benefit from the Ghana School Feeding Programme (SFP) but were not yet enrolled, were
selected as control schools based on their similarity with the SFP pilot schools with respect to
the following characteristics: number of children enrolled in school; school infrastructure; size
of community; absence of market infrastructure; water and sanitation facilities and proximity
to each other [
]. Subsequently, children who met the inclusion criteria were randomly
assigned to the treatment and control arms of the dietary trial; details of the selection
procedure have been published elsewhere [
We assessed the dietary diversity score (DDS) of the SAC at two different time points using
a qualitative 24-hour dietary recall (24hR). The baseline assessment (October 2010) was
conducted before the intervention and corresponded with the onset of the dry season and the
harvest of most crops especially cereals, legumes like cowpea and some root tubers particularly
sweet potatoes. On the contrary, the follow-up assessment (May 2011) was conducted at the
end of the intervention and was in the rainy season coinciding with the lean period when
many households have their staple food (cereal and root tubers) stocks depleted.
At baseline, anthropometric measurements (weight and height) of the children were taken
and a pretested semi-structured questionnaire was used to assess the socio-demographic and
economic characteristics of the children and their households. Some children were lost to
follow-up (n = 13) and only children with data at both time points (n = 228) were included in the
final analysis. Ethical approval for the study was obtained from the Institutional Review Board
of Noguchi Memorial Institute for Medical Research, University of Ghana and Medical
Research Ethics Committee of Wageningen University, The Netherlands. We obtained
permission from the district administration, chiefs and opinion leaders of the respective
communities. Thumb-printed informed consent was also obtained from each parent or caregiver.
A qualitative 24hR was used to assess the dietary intake of the SAC at both time points
(October 2010 and May 2011). Although the SAC were the primary subjects, mothers and caregivers
were respondents for the children as children are likely to give inaccurate descriptions of their
dietary intake [
]. Mothers and caregivers were first asked to mention all foods including
drinks and snacks that were consumed the previous 24 hours (from wake-up to wake-up)
preceding the survey by the index child from home and outside of the home. She was next probed
for likely forgotten foods and then asked to give a detailed description of foods and beverages
consumed, including ingredients for mixed dishes. To ensure intake outside the home was
captured, children were asked to assist their mothers/caregivers in the recall.
3 / 16
Dietary diversity score (DDS)
DDS was calculated for each child based on a set of 13 food groups namely: (1) starchy staples,
(2) legumes and nuts, (3) dairy products, (4) organ meat, (5) eggs, (6) small whole fish, (7)
flesh foods, (8) vitamin A rich dark green leafy vegetables, (9) vitamin A-rich deep yellow,
orange and red vegetables, (10) vitamin A-rich fruits (vitamin A
120RE/100g), (11) vitamin
C-rich vegetables (vitamin C
9 mg/100g), (12) vitamin C-rich fruits (vitamin C
100g) and (13) all other fruits and vegetables. The choice of the food groups was based on
Arimond et al [
]. Definition of Vitamin A rich and vitamin C-rich fruits or vegetables was
based on the FAO definition [
]; subsequently based on the nutrient content of the vegetables
or fruits in the Ghana/Mali Food Composition Tables [
], they were classified accordingly
as vitamin A rich dark green leafy vegetables, vitamin A-rich fruits and vitamin A rich-deep
yellow orange and red vegetables. The food items in each food group are presented in Table 1.
At baseline and follow-up, a score of 1 was assigned if a child consumed a food item
belonging to a particular food group, else 0. Individual food group scores were aggregated into DDS
for each child at baseline and follow-up. Any food consumed on multiple occasions during the
24 hour period was counted only once resulting in a maximum attainable score of 13. The
scoring did not consider a minimum intake for the food groups.
We also created a score for vitamin A rich fruits and/or vegetables consumption (sum score
for: vitamin A-rich dark green leafy vegetables, vitamin A-rich deep yellow, orange and red
vegetables and vitamin A-rich fruits), vitamin C-rich fruits/vegetables consumption (sum
score for: vitamin C-rich vegetables and vitamin C-rich fruits) and animal source foods (ASF)
4 / 16
consumption (sum score for: all dairy products, organ meat, small fish eaten whole, eggs, and
all other flesh foods). The score for vitamin A-rich fruits and/or vegetable intake, vitamin
Crich fruits and/or vegetable intake and ASF intake ranged from 0±3, 0±2 and 0±5 respectively.
The weight and height of the children were measured following standard procedures [
Weight was measured to the nearest 0.1kg with a calibrated electronic scale (UNIscale; Seca)
whilst standing height was measured to the nearest 0.1 cm with a microtoise (Bodymeter 208;
Seca). The measurements were duplicated and averaged to reduce random instrumental error.
We used the anthropometry to compute the Body Mass Index-for-age z-score (BAZ),
weightfor age z-score (WAZ) and height-for-age Z-score (HAZ) of each child. The completed age of
each child in years was determined using any reliable source (birth certificate, child health
record, health insurance card) and the date of collection of anthropometric data. Where there
was no reliable source of birth date, the parents/caregiver were asked to estimate age based on
another child's records or event on the traditional calendar (12 children) [
We assessed the demographic and socioeconomic characteristics of the children's
households with a pretested semi-structured questionnaire. Demographic and socioeconomic
related covariates included child's compound size (number of people living in child's house),
educational status of household head and mother as well as the occupation of household head
Population characteristics were presented as means (standard deviations) for continuous
variables and frequency (percentages) for categorical variables. Univariate analysis of variance
with the MIXED procedure [
] was used to analyze the association between DDS and
seasonality. Potential confounders included as covariates were a priori selected based on literature
and included age, sex, nutritional status of the child (underweight), total compound size,
educational status of household head and mother as well as the occupation of the household head
and mother [17±19,39]. Spearman and Pearson correlation coefficient showed no
multicollinearity between the confounding variables. The model was checked for effect modification by
sex and age but none was statistically significant and were thus excluded in the final analysis.
Besides the crude model, we created 2 multivariate models. The first multivariate model was
adjusted for demographic factors (age and sex). The full model was further adjusted for
socioeconomic factors of child's household (educational status of household head, the occupation
of household head and occupation of the mother). Whilst occupation of household head was
modelled as farmer or other (trader and civil servant), maternal occupation was modelled as
farmer, trader and other (housewife and civil servant) based on the data available. Although a
priori chosen, we excluded nutritional status defined by BAZ and total compound size from
the models as they had no influence on the results. Furthermore, educational status of mother
and household head had low variability; we, however, modelled the educational status (literate
or non-literate) of the household head which had more variability than that of the mother.
We also compared the DDS at baseline and end-line between the treatment and control
arms of the feeding trial using One-Way ANOVA but found no differences between treatment
and control arms of the feeding trial; hence we did not adjust for the intervention effect. The
X2 statistic was used to determine the association between the percentage of children
consuming a food group and seasonality as well as the association between DDS category (low and
high) and seasonality. Although DDS is commonly categorized into tertiles (low, moderate
and high), the median split (low and high) was preferred because there was little variation
5 / 16
(DDS difference = 1) between the 33rd percentile (DDS = 6) and the 67th percentile (DDS = 7);
even when using the tertile categorization similar to the FAO (DDS 3, DDS of 4±5 and DDS
], only 2 DDS categories similar to the median split were created. All statistical analyses
were done using SAS 9.3 (SAS Institute Inc., Cary NC.) and a two-tailed P-value 0.05 at 95%
confidence interval was considered statistically significant.
Characteristics of the sample
Table 2 shows the characteristics of the study population at baseline. The mean age of the SAC
was 8.1 ± 2.1 years and the mean BAZ was -0.5 ± 0.8. About 62.7% of the children were boys
and the average compound size was 15.5 ± 8.7. The household heads (91.1%), as well as the
mothers (69.3%), were predominantly farmers. The majority of the household heads (84.1%)
and the mothers/caregivers (97.8%) of the children had never been to school.
6 / 16
Food consumption pattern and seasonality
Except for small fish eaten whole, < 10% each of the children consumed any other ASF food
group in either season (Table 3). When, excluding small fish eaten whole, the percentage of
children consuming at least one ASF in the dry season was more than twice the percentage of
children consuming it in the rainy season (13.2% vs 4.8%, P = 0.002). Compared to the dry
season, more children consumed vitamin A-rich dark green leafy vegetables (52.6% vs. 23.3%,
P < .0001) and vitamin A-rich fruits (64.0% vs. 0.9%, P < .0001) in the rainy season.
Conversely, more children consumed vitamin A-rich deep yellow, orange and red vegetables in the
dry season compared to the rainy season (73.7% vs. 36.4%, P < .0001). However, the
percentage of children consuming at least one vitamin A-rich fruit/vegetable and vitamin C-rich fruit/
vegetable were each significantly higher in the rainy season compared to the dry season (90.7%
vs. 81.1%, P = 0.003 and 100.0% vs. 96.5%, P = 0.004 respectively).
The consumption of legumes and nuts increased significantly in the rainy season compared
to the dry season (98.3% vs. 89.9%, P = 0.0002) whilst the consumption of other fruits and
vegetables was marginally higher in the dry season compared to the rainy season (93.4% vs.
90.8%, P = 0.30). Furthermore, the consumption of vitamin C-rich fruits was poor in both
seasons but vitamin C-rich vegetables were consumed by all the children in the rainy season
compared to the dry season (100% vs. 96.5%, P = 0.004). Finally, all children consumed starchy
staples in both seasons.
When categorizing DDS into low and high, more children had a high DDS in the rainy
season compared to the dry season (88.2% vs. 72.4%, P < .0001) and the percentage of children
with low DDS in the dry season was more than twice the percentage in the rainy season (27.6%
vs. 11.8%) (Table 4).
1 Chi-square test of independence was used to determine the association between the percentage of children consuming a food group and seasonality.
2Excludes small whole fish but includes dairy products, organ meat, eggs, and all other flesh foods
*P-value statistically significant at 5% level of significance
7 / 16
1 Chi-square test of independence was used to determine the association between the percentage of children with low or high DDS and seasonality
*P-value statistically significant at 5% level of significance.
Association between DDS and seasonality
The mean DDS of the children was significantly higher in the rainy season compared to the
dry season in both crude and multivariate models (Table 5). Adjustment for demographic
factors (sex and age) made no difference in the observed means between seasons. Compared to
the crude mode, the mean DDS for both seasons increased marginally in the full multivariate
model. However, the difference in means between seasons decreased marginally in the full
multivariate model compared to the crude model, with the mean DDS remaining significantly
higher in the rainy season compared to the dry season (6.54 ± 0.09 vs. 6.00 ± 0.09, P < .0001).
None of the household covariates was significantly associated (P >0.05) with the DDS of the
children in either seasons (S1 Table).
From Table 6, the mean score for the consumption of vitamin A-rich fruits and/or
vegetables was significantly higher in the rainy season compared to the dry season (1.56 ± 0.07 vs.
1.01 ± 0.07, P < .0001). On the contrary, the mean score for consumption of ASF was
significantly higher in the dry season compared to the rainy season (1.11 ± 0.03 vs. 1.03 ± 0.03,
P = 0.008). Lastly, the mean score for the consumption of vitamin C-rich fruits and/or
vegetables was similar for both the dry and rainy seasons (1.01 ± 0.02 vs. 0.98 ± 0.02, P = 0.17).
In this study, we evaluated the effect of seasonality on the DDS of SAC in rural northern
Ghana. In fact, the children had significantly more diverse diets in the rainy season than the
dry season, with dietary diversity increasing from a mean of 6.00 food groups (out of 13) in the
dry season to 6.54 in the rainy season. Indeed, other studies have also shown food
consumption can vary from one season to the next [20±23] and this may directly affect the diversity of
Model 1 was adjusted for age and sex; Model 2 was further adjusted for socioeconomic factors of child's household: educational level of household head,
the occupation of household head and maternal occupation.
1Mean DDS in the rainy season minus mean DDS in the dry season
*P-value is significant at 5% level of significance.
8 / 16
SE, standard error; T1, dry season; T2 = rainy season;
Model 1 was adjusted for age and sex; Model 2 was further adjusted for socioeconomic factors of child's household: educational level of household head,
the occupation of the household head and maternal occupation
*P-value is significant at 5% level of significance.
diet, especially in the rural Africa setting. Seasonal variations in nutrient intake signifying
seasonal differences in food consumption have also been documented in Burkina Faso [
Our present results are similar to those of Savy et al. [
] who found an increase in DDS
from 3.4 ± 1.1 to 3.8 ± 1.5 between the beginning and the end of the food shortage season
(P < 0.0001) in a sample of Burkinabe women.
Generally, the consumption of vitamin C-rich fruits was poor in both seasons with < 2% of
the children consuming any vitamin C-rich fruit in either of the seasons. Studies suggest an
intake of fruits and snacks outside the home by SAC is more likely to be omitted by the
mothers and caregivers during dietary recalls such as used in our survey [
]. Although this may
partly explain the poor consumption of vitamin C-rich fruits in both seasons, the children
assisted their mothers/caregivers recall their intake outside the home. It was, therefore, less
likely that under reporting of fruit intake influenced our results.
We found that a significant proportion of the children consumed vitamin A-rich dark
green leafy vegetables and vitamin A-rich fruits in the rainy season than the dry season.
Remarkably, < 1% of the children consumed vitamin A-rich fruits in the dry season; this
difference may partly be attributed to the intake of shea fruits and mangoes, available only in the
early parts of the rainy season. The consumption of vitamin C-rich vegetables was high in both
seasons; however, all children consumed vitamin C-rich vegetables in the rainy compared to
the dry season. On the contrary, the consumption of vitamin A-rich deep yellow, orange and
red vegetables was significantly higher in the dry season compared to the rainy season and
may partly be attributed to the availability of orange flesh sweet potatoes which is seasonal and
was been harvested at the time of the baseline assessment (October).
Overall, the difference in mean DDS between seasons was mainly due to the difference in
consumption of fruits and vegetables between the seasons. Out of 6 food groups which differed
significantly between the rainy season and dry season, 4 food groups could broadly be
classified as fruits and vegetables with 3 of them significantly higher in the rainy season than the dry
season. Additionally, of the 6 food groups which differed significantly between the seasons, 3
9 / 16
were vitamin A-rich fruits/vegetables and 2 of these were significantly higher in the rainy
season compared to the dry season. Recent studies from Nigeria [
] and Burkina Faso [
found that fruit and/or green leafy vegetable consumption was higher in the rainy season
compared to the dry season. In the present study, the difference in consumption of fruits and
vegetables between seasons may partly be explained by the seasonal availability of fruits and
vegetables in Ghana. A study in Ghana confirmed seasonality influences the consumption of
vitamin A-rich foods as vegetables and fruits were found to be more abundant in the rainy
season than the dry season [
]. In our experience, the prices of fresh fruits and vegetables are
generally higher in the dry season compared to the rainy season, implying poor access for
deprived households during the dry season; this may partly account for the differences
between the seasons.
Although the results showed that a majority of the children consumed at least one vitamin
A-rich fruit and/or vegetable in both seasons, the percentage of children consuming at least
one vitamin A-rich fruit and/or vegetable was significantly higher in the rainy season
compared to the dry season. Moreover, the mean score for the consumption of vitamin A-rich
fruits and/or vegetables was significantly higher in the rainy season compared to the dry season
(mean score difference = 0.55 ± 0.07, P < .0001). In the present study, the mean score
difference between seasons for vitamin A-rich fruits/vegetables was similar to the overall DDS mean
difference between seasons (0.55 ± 0.07 vs.0.54 ± 0.08); emphasizing that the difference in
mean DDS between seasons was mainly due to the difference in consumption of vitamin
Arich fruits and/or vegetables. A recent study in Bangladesh [
] also reported that differences
in DDS by season corresponded with differences in consumption of dark green leafy
vegetables and vitamin A-rich fruits/vegetables.
The present analysis did not ascertain whether or not the statistically significant difference
in the mean DDS between seasons or the percentage of children consuming each food group is
of nutritional importance. Nonetheless, Savy et al [
] showed that a difference in mean DDS
between seasons (0.55) similar to that found in our study, has an effect on nutritional status of
women in rural Burkina Faso. In another study of younger children (aged 12±59 months) in
rural Burkina Faso, Arsenault et al. [
] reported that a difference in mean DDS (mean
difference = 0.24) between the pre and post-harvest seasons was associated with a higher
micronutrient intake in the post-harvest season compared to the pre-harvest season. Furthermore,
Nupo et al. [
] found that a 0.38 difference in mean DDS between seasons is associated with a
significant difference in energy, protein and fat intake between seasons in a group of rural
Nigerian women. Although the difference in mean DDS across seasons in the present study
directly illustrates fluctuations in dietary diversity across seasons among the SAC, it is plausible
the difference in mean DDS and/or consumption of vitamin A-rich fruits/vegetables between
seasons may influence nutrient intake and consequently dietary adequacy between seasons;
but, further studies are needed to establish this.
Adjustment for potential confounders such as age, sex and socio-demographic factors
(educational status of household head and occupation of mother/household head) did not change
the observed difference between the two seasons. Several studies have shown that household
food security and DDS improve with maternal and/or paternal education [
it has also been shown that a household head or mother with a better paying job improves
]. Although we did not model maternal education due to its homogeneity in the
present study, none of the modelled household covariates (education of household head, the
occupation of mother and occupation of household head) was significantly associated with the
DDS of the children; perhaps a result of the homogeneity of the sample.
In conformity with other studies [22,30,46±48], the results showed poor consumption of
ASF in both seasons. Besides ªsmall fish eaten wholeº which was consumed by almost every
10 / 16
child in both seasons, < 10% each of the children consumed any other ASF in either of the
seasons. That withstanding, small fish eaten whole is often consumed in minute quantities in
soups and stews in Ghana and may thus be insignificant in improving micronutrient
adequacy. Ruel [
] and unpublished data from Ghana confirm small fish eaten whole (anchovies)
contributes very little to micronutrient adequacy [
]. This low consumption of ASF in both
seasons may have serious consequences for the micronutrient adequacy of the children in
either of the seasons as ASF are known to better improve micronutrient adequacy especially
for Fe and Zn [50±52] than plant source foods which have poor bioavailability for Fe and Zn
[53±55]. In fact, ASF are generally among the most expensive foods [
] in Ghana and may be
the reason for their poor consumption in our study population. However, the design of the
present study limited its ability to evaluate the effect of the poor ASF intake on the
micronutrient adequacy of the children. Similar to other studies [
], staples were equally important
foods for the children in both seasons as every child consumed a staple (cereals and grains as
well as starchy roots and tubers) in both seasons. Even though legumes and nuts were a part of
the diet of most of the children in both seasons, the consumption of legumes and nuts was
significantly higher in the rainy season compared to the dry season. Liere et al , reported that
consumption of legumes increases when cereal stocks are becoming depleted; accordingly, the
higher legume consumption in the rainy season compared to the dry season may signal a
reduction in the quantity of available household cereal stocks. Some studies have reported
legumes as good sources of plant protein, with fairly good amounts of Fe and Zn [
Nonetheless, the current study could not determine whether or not the statistically significant
difference in consumption of legumes between seasons is of nutritional importance as we did
not assess nutrient intake or dietary adequacy in the present study.
Our study highlights the importance of measuring DDS across the year in order to identify
seasonal variations in dietary intake. As a result, researchers who undertake dietary surveys in
a context similar to ours need to be mindful of the importance of including data from different
seasons. Interventions that aim to improve the diet of young children from low-income, rural
communities need to recognize the role of seasonality on diet and to incorporate initiatives to
prevent seasonal declines. In designing context-specific interventions to improve the diet of
young children, policy makers and programme planners also need to incorporate initiatives
that prevent undesirable seasonal declines in dietary diversity.
Likewise, in the design of pre and post-test nutrition evaluation programmes, the effect of
seasonality needs to be considered; Swindale and Bilinsky [
] recommended repeated data
collection should be undertaken at the same time of the year to avoid seasonal differences. Our
findings, therefore, corroborate their recommendation as we have shown that DDS changes
across the seasons.
Several strengths and limitations should be taken into account when interpreting the
findings of the present study. Firstly, the main strength of the current study lies in its longitudinal
design which allowed us to make a comparison of DDS from two seasons among the same
individuals. Secondly, the quality of our data is high as we pretested our questionnaire and
skilled interviewers who were given a 5-day training undertook the data collection.
On the contrary, residual confounding is always possible in observational studies even with
extensive statistical adjustments [
]. Hence, though we adjusted for several potential
confounders in this study, we cannot rule out the effect of residual confounding in our present
Notably, seasonality is known to influence household food security [
] which is a
key determinant of household dietary diversity as food insecure households often consume
low-quality monotonous diets which are low in micronutrients [
], but the present study
did not include any measure of food security. Nonetheless, food security is well associated with
11 / 16
household demographic and socioeconomic characteristics including occupation [
] and household size [
]; thus our adjustment for education, occupation
and child compound size may have partly accounted for the effect of household food security.
Our use of a single 24hR and the lack of portion sizes to include a minimum intake for
scoring is the main limitation of our dietary assessment method. A single 24hR may not
reflect habitual dietary intake and may capture episodic foods that are not typically consumed
]. There is also evidence that qualitative recalls lead to over reporting and consequently
higher scores and misclassification of subjects [
]. Thus, it is possible the DDS for both
seasons could have been lower than we found if a minimum intake was set for the DDS.
More so, the qualitative assessment limited the ability of the present study to determine if the
observed difference in mean DDS between seasons reflected seasonal differences in dietary
adequacy, especially given that vitamin A-rich foods and legumes were consumed by most
children in both seasons and ASF consumption in both seasons was generally poor. On the
contrary, the qualitative 24hR was less subject to recall bias, less burdensome to respondents,
fast, easy and also conforms to the recall time period used in many dietary diversity studies
[2,14,16±18]. Although a food frequency questionnaire is ideal for assessing habitual dietary
intake, it is generally associated with more recall bias compared to a 24hR [
]; thus, a
repeated 24hR may be more appropriate in capturing habitual diet.
The end-line assessment was conducted at the end of the feeding trial, it was therefore
unlikely that the daily provision of 430kcal in the intervention trial [
] reduced dietary intake
in the end-line. However, the children were treated for helminths infestation after the baseline
assessment and 3.5 months into the intervention which could have influenced appetite in the
end-line compared to the non-intervention situation. Generally, randomization of children
into treatment or control groups had no effect on the DDS at baseline and end-line (results
not shown). Caution is however needed in extending the results presented here to larger
contexts; at best, the present findings can be extrapolated to rural school-aged children in
Northern Ghana where culture and dietary patterns are quite similar.
In summary, the results of the present study indicate seasonality has an effect on the food
consumption pattern and dietary diversity of school-age children in rural Northern Ghana.
Although the present study did not evaluate seasonal variability in dietary adequacy, DDS is
known to be a proxy indicator of micronutrient intake; thus the seasonal differences in DDS
may have an influence on the micronutrient intake of the school-age children between
S1 Table. Association of household factors with dietary diversity (DDS) for school-aged
children in Tolon district by season. DDS, dietary diversity; SE, standard error of mean;
1P-value for interaction between household factor and season; for each household factor,
Means ± SE with similar superscripts within column do not differ significantly at 5% level of
S1 Data. Dataset for dietary diversity and seasonality.
12 / 16
The authors are grateful to the teachers of the 2 schools: Y. Abdul-Majeed, A. Alaru, A.A.
Suhuyini, and A. Wumbei. The authors also thank the mothers/caregivers who participated in
Conceptualization: Abdul-Razak Abizari, Inge D. Brouwer.
Formal analysis: Abdul-Razak Abizari, Fusta Azupogo.
Funding acquisition: Abdul-Razak Abizari.
Investigation: Abdul-Razak Abizari, Miwako Nagasu, Noortje Creemers.
Methodology: Abdul-Razak Abizari, Inge D. Brouwer.
Supervision: Abdul-Razak Abizari, Inge D. Brouwer.
Writing ± original draft: Abdul-Razak Abizari, Fusta Azupogo, Miwako Nagasu, Noortje
Writing ± review & editing: Abdul-Razak Abizari, Fusta Azupogo, Inge D. Brouwer.
13 / 16
14 / 16
15 / 16
1. Hatløy A , Torheim LE , Oshaug A . Food variety-a good indicator of nutritional adequacy of the diet ? A case study from an urban area in Mali, West Africa . Eur J Clin Nutr . 1998 ; 52 : 891 ± 898 . https://doi.org/ 10.1038/sj.ejcn. 1600662 PMID: 9881884
2. Arimond M , Wiesmann D , Becquey E , Carriquiry A , Daniels MC , Deitchler M , et al. Simple Food Group Diversity Indicators Predict Micronutrient Adequacy of Women ` s Diets in 5 diverse, resource-poor settings . J Nutr . 2010 ; 140 : 2059s± 2069s . https://doi.org/10.3945/jn.110.123414 PMID: 20881077
3. Ruel MT . Is dietary diversity an indicator of food security or dietary quality? A review of measurement issues and research needs . Food and Nutrition Bulletin . 2003 .
4. Kant AK . Indexes of overall diet quality: A review . J Am Diet Assoc . 1996 ; 96 : 785 ± 791 . https://doi.org/ 10.1016/S0002- 8223 ( 96 ) 00217 - 9 PMID: 8683010
5. Kant aK , Thompson F. Measures of overall diet quality from a food frequency questionnaire: National Health Interview Survey . Nutr Res . 1997 ; 17 : 1443 ± 1456 .
6. Drewnowski A , Renderson SA , Driscoll A , Rolls BJ . The Dietary Variety Score . J Am Diet Assoc . 1997 ; 97 : 266 ± 271 . https://doi.org/10.1016/S0002- 8223 ( 97 ) 00070 - 9 PMID: 9060943
7. LoÈwik MR , Hulshof KF , Brussaard JH . Food-based dietary guidelines: some assumptions tested for The Netherlands . Br J Nutr . 1999 ; 81 Suppl 2 : S143 ± S149 . https://doi.org/10.1017/ S0007114599001002
8. Ogle BM , Hung PH , Tuyet HT . Significance of wild vegetables in micronutrient intakes of women in Vietnam: an analysis of food variety . Asia Pac J Clin Nutr . 2001 ; 10 : 21 ± 30 . https://doi.org/10.1046/j.1440- 6047 . 2001 . 00206 . x PMID : 11708605
9. Torheim LE , Barikmo I , Parr CL , Hatløy a , Ouattara F , Oshaug a. Validation of food variety as an indicator of diet quality assessed with a food frequency questionnaire for Western Mali . Eur J Clin Nutr . 2003 ; 57 : 1283 ± 1291 . https://doi.org/10.1038/sj.ejcn. 1601686 PMID: 14506490
10. Steyn NP , Nel J , Labadarios D , Maunder EMW , Kruger HS . Which dietary diversity indicator is best to assess micronutrient adequacy in children 1 to 9 y? Nutrition . Elsevier Inc.; 2014 ; 30 : 55 ± 60 . https://doi. org/10.1016/j.nut. 2013 . 06 .002 PMID: 24290599
11. Kennedy GL , Pedro MR , Seghieri C , Nantel G , Brouwer I. Dietary diversity score is a useful indicator of micronutrient intake in non-breast-feeding Filipino children . J Nutr . 2007 ; 137 : 472 ± 477 . PMID: 17237329
12. Moursi MM , Arimond M , Dewey KG , Trèche S , Ruel MT , Delpeuch F . Dietary diversity is a good predictor of the micronutrient density of the diet of 6- to 23-month-old children in Madagascar . J Nutr . 2008 ; 138 : 2448 ± 2453 . https://doi.org/10.3945/jn.108.093971 PMID: 19022971
13. Moursi MM , Martin-PreÂvel Y , Eymard-Duvernay S , Capon G , Trèche S , Maire B , et al. Assessment of child feeding practices using a summary index: Stability over time and association with child growth in urban Madagascar . Am J Clin Nutr . 2008 ; 87 : 1472 ± 1479 . PMID: 18469273
14. Arimond M , Ruel MT . Dietary Diversity Is Associated with Child Nutritional Status : Evidence from 11 Demographic and Health Surveys . J Nutr . 2004 ; 134 : 2579 ± 2585 . PMID: 15465751
15. Hatløy A , Hallund J , Diarra MM , Oshaug A . Food variety, socioeconomic status and nutritional status in urban and rural areas in Koutiala (Mali) . Public Health Nutr . 2000 ; 3 : 57 ± 65 . PMID: 10786724
16. Daniels MC , Adair LS , Popkin BM , Truong YK . Dietary diversity scores can be improved through the use of portion requirements: an analysis in young Filipino children . Eur J Clin Nutr . 2009 ; 63 : 199 ± 208 . https://doi.org/10.1038/sj.ejcn. 1602927 PMID: 17971828
17. Savy M , Martin-Prevel Y , Sawadogo P , Kameli Y , Delpeuch F . Use of variety / diversity scores for diet quality measurement : relation with nutritional status of women in a rural area in Burkina Faso . Eur J Clin Nutr . 2005 ; 59 : 703 ± 716 . https://doi.org/10.1038/sj.ejcn. 1602135 PMID: 15867942
18. Savy M , Martin-PreÂvel Y , Traissac P , Eymard-Duvernay S , Delpeuch F. Dietary diversity scores and nutritional status of women change during the seasonal food shortage in rural Burkina Faso . J Nutr . 2006 ; 136 : 2625 ± 2632 . PMID: 16988137
19. Savy M , Martin-PreÂvel Y , Traissac P , Delpeuch F . Measuring dietary diversity in rural Burkina Faso: comparison of a 1-day and a 3-day dietary recall . Public Health Nutr . 2007 ; 10 : 71 ± 78 . https://doi.org/ 10.1017/S1368980007219627 PMID: 17212836
20. Liere MJ van , Ategbo E-AD , Hartog AP Den, Hautvast JGAJ . The consequences of seasonal food insecurity for individual food-consumption patterns in north-western Benin . Food Nutr Bull . 1995 ; 16 : 147 ± 54 .
21. Tetens I , Hels O , Khan NI , Thilsted SH , Hassan N. Rice-based diets in rural Bangladesh: How do different age and sex groups adapt to seasonal changes in energy intake? Am J Clin Nutr . 2003 ; 78 : 406 ± 413 . PMID: 12936922
22. Mitchikpe CES , Dossa R a M , Ategbo E a D, Van Raaij JM a, Kok FJ . Seasonal variation in food pattern but not in energy and nutrient intakes of rural Beninese school-aged children . Public Health Nutr . 2009 ; 12 : 414 ± 422 . https://doi.org/10.1017/S1368980008002929 PMID: 18616847
23. Ferguson EL , Gibson RS , Opare-Obisaw C , Osei-Opare F , Lamba C , Ounpuu S . Food consumption patterns and dietary diversity of rural preschool Ghanaian and Malawian children . Ecol Food Nutr . Gordon and Breach Science Publishers; 1993 ; 29 : 219 ± 234 .
24. Hudelson P , Dzikunu H , Mensah JD , Morris S , Kirkwood B . Dietary patterns in a rural area of Ghana and their relevance for vitamin A consumption . Ecol Food Nutr . 1999 ; 38 : 83 ± 20 . https://doi.org/10. 1080/03670244. 1999 .9991577
25. M'Kaibi FK , Steyn NP , Ochola S , Du Plessis L. Effects of agricultural biodiversity and seasonal rain on dietary adequacy and household food security in rural areas of Kenya. BMC Public Health .???; 2015 ; 15 : 1± 11 .
26. Becquey E , Delpeuch F , KonateÂ AM , Delsol H , Lange M , Zoungrana M , et al. Seasonality of the dietary dimension of household food security in urban Burkina Faso . Br J Nutr . 2012 ; 107 : 1860 ± 1870 . https:// doi.org/10.1017/S0007114511005071 PMID: 22017887
27. Arsenault JE , Nikiema L , Allemand P , Ayassou KA , Lanou H , Moursi M , et al. Seasonal differences in food and nutrient intakes among young children and their mothers in rural Burkina Faso . J Nutr Sci . 2014 ; 3: 1±9 . https://doi.org/10.1017/jns. 2014 .53 PMID: 26101623
28. Ghana districts. A repository of all districts in the Republic of Ghana . In: Northern » Tolon [Internet]. 2006 [cited 20 Nov 2016 ]. http://ghanadistricts.gov.gh/About-District-Details.aspx?distID= 141 &distName=Tolon
29. Abizari A , Moretti D , Zimmermann MB , Armar-Klemesu M , Brouwer ID . Whole Cowpea Meal Fortified with NaFeEDTA Reduces Iron Deficiency among Ghanaian School Children in a Malaria Endemic Area . J Nutr . 2012 ; 142 : 1836 ± 1842 . https://doi.org/10.3945/jn.112.165753 PMID: 22915294
30. Abizari A-R , Buxton C , Kwara L , Mensah-Homiah J , Armar-Klemesu M , Brouwer ID . School feeding contributes to micronutrient adequacy of Ghanaian schoolchildren . Br J Nutr . 2014 ; 112 : 1019 ± 1033 . https://doi.org/10.1017/S0007114514001585 PMID: 24990068
31. Livingstone MB , Robson PJ . Measurement of dietary intake in children . Proc Nutr Soc . 2000 ; 59 : 279 ± 93 . PMID: 10946797
32. McPherson RS , Hoelscher DM , Alexander M , Scanlon KS , Serdula MK . Dietary Assessment Methods among School-Aged Children: Validity and Reliability . Prev Med (Baltim) . 2000 ; 31 : S11±S33 . https:// doi.org/10.1006/pmed. 2000 .0631
33. Arimond M , Wiesmann D , Becquey E , Carriquiry A , Daniels M , Deitchler M , et al. Dietary diversity as a measure of the micronutrient adequacy of women's diets in resource-poor areas: summary of results from five sites . Washington D.C: FHI 360; 2011 .
34. FAO . Guidelines for measuring household and individual dietary diversity . Rome, Italy; 2011 .
35. Barikmo I , Quattara F , Oshaug A . Table de composition d'aliments du Mali, Composition Table for Mali . Oslo, Norway; 2004 .
36. Barreau NZ . Food Composition Table for use in Ghana . Accra, Ghana; 1992 .
37. WHO. WHO Child Growth Standards. Length/height-for-age, weight-for-age, weight-for-length, weightfor-height and body mass index-for-age. Methods and development . Geneva: WHO press; 2006 .
38. Littell RC , Milliken GA , Stroup WW , Wolfinger RD , Schabenberger O. SAS for Mixed Models . second. Cary, NC SAS Institute Inc.; 2006 .
39. Harris-Fry H , Azad K , Kuddus A , Shaha S , Nahar B , Hossen M , et al. Socio-economic determinants of household food security and women's dietary diversity in rural Bangladesh: a cross-sectional study . J Heal Popul Nutr. Journal of Health, Population and Nutrition; 2015 ; 33 :  p. http ://dx.doi.org/10.1186/ s41043-015-0022-0
40. Ferguson EL , Gibson RS , Opare-Obisaw C. The relative validity of the repeated 24h recall for estimating energy and selected nutrient intakes o rural Ghanaian children . Eur J Clin Nutr . 1994 ; 48 : 241 ± 252 . PMID: 8039484
41. Gewa C a , Murphy SP , Neumann CG . Out-of-home food intake is often omitted from mothers' recalls of school children's intake in rural Kenya . J Nutr . 2007 ; 137 : 2154 ± 2159 . 137/9/2154 [pii] PMID: 17709457
42. Nupo SS , Oguntona CRB , Onabanjo OO , Fakoya EO . Dietary diversity scores and nutritional status of women in two seasons in rural areas of Ogun State, Nigeria . Nutr Food Sci. 2013 ; 43 : 60 ± 67 . https:// doi.org/10.1108/00346651311295923
43. Stevens B , Watt K , Brimbecombe J , Clough A , Judd J , Lindsay D. The role of seasonality on the diet and household food security of pregnant women living in rural Bangladesh : a cross-sectional study . Public Health Nutr . 2016 ; 1±9 . https://doi.org/10.1017/S136898001600183X PMID: 27573667
44. Hillbruner C , Egan R. Seasonality , household food security, and nutritional status in Dinajpur, Bangladesh . Food Nutr Bull. 2008 ; 29 : 221 ± 231 . https://doi.org/10.1177/156482650802900308 PMID: 18947035
45. Faber M , Schwabe C , Drimie S . Dietary diversity in relation to other household food security indicators . Int J Food Safety, Nutr Public Heal . 2009 ; 2 : 1± 15 .
46. Tatala S , Ndossi G , Ash D. Impact of dietary iron intake on anaemia in Tanzanian schoolchildren . SAJCN . 2004 ; 17 : 94 ± 100 .
47. Gewa C a , Murphy SP , Weiss RE , Neumann CG . Determining minimum food intake amounts for diet diversity scores to maximize associations with nutrient adequacy: an analysis of schoolchildren's diets in rural Kenya . Public Health Nutr . 2014 ; 17 : 1± 10 .
48. Tidemann-Andersen I , Acham H , Maage A , Malde MK . Iron and zinc content of selected foods in the diet of schoolchildren in Kumi district, east of Uganda: a cross-sectional study . Nutr J. BioMed Central Ltd; 2011 ; 10 : 81 . https://doi.org/10.1186/ 1475 -2891-10-81 PMID: 21827701
49. Azupogo F. Development of food-based recommendations for 6±8 months old breastfed children in Karaga District-Ghana using linear programming . Wageningen University and Research Centre, Unpublished . 2015 .
50. Cleghorn G . Role of red meat in the diet for children and adolescents . Nutr Diet . 2007 ; 64 : s143± s146 . https://doi.org/10.1111/j.1747- 0080 . 2007 . 00203 .x
51. Neumann CG , Bwibo NO , Murphy SP , Sigman M , Whaley S , Allen LH , et al. Animal Source Foods Improve Dietary Quality, Micronutrient Status, Growth and Cognitive Function in Kenyan School Children : Background, Study Design and Baseline Findings . J Nutr . 2003 ; 133 : 3941 ± 3949 .
52. Pelto GH , Armar-Klemesu M , Siekmann J , Schofield D. The focused ethnographic study ªassessing the behavioral and local market environment for improving the diets of infants and young children 6 to 23 months oldº and its use in three countries . Matern Child Nutr . 2013 ; 9 : 35 ± 46 . https://doi.org/10.1111/j. 1740- 8709 . 2012 . 00451 . x PMID : 23167583
53. Hambrñus L. ªMeat or wheat for the next millennium?º Plenary Lecture. Animal- and plant-food-based diets and iron status : benefits and costs . Proc Nutr Soc . 1999 ; 58 : 235 ± 242 .
54. Gibson RS , Bailey KB , Gibbs M , Ferguson EL . A review of phytate, iron, zinc, and calcium concentrations in plant-based complementary foods used in low-income countries and implications for bioavailability . Food Nutr Bull . 2015 ; 31 : 134 ± 146 .
55. Gibson RS , Perlas L , Hotz C . Improving the bioavailability of nutrients in plant foods at the household level . Proc Nutr Soc . 2006 ; 65 : 160 ± 168 . https://doi.org/10.1079/PNS2006489 PMID: 16672077
56. Colecraft E , Marquis GS , Aryeetey R , Sakyi-Dawson O , Lartey A , Ahunu B , et al. Constraints on the Use of Animal Source Foods for Young Children in Ghana: A Participatory Rapid Appraisal Approach . Ecol Food Nutr . 2006 ; 45 : 351 ± 377 . https://doi.org/10.1080/03670240600985464
57. Tanner M , Lukmanji Z . Food consumption patterns in a rural Tanzanian community (Kikwawila village, Kilombero District, Morogoro Region) during lean and post-harvest season . Acta Trop . 1987 ; 44 : 229 ± 244 . PMID: 2891271
58. Liere MJ Van , Ategbo E , Den Hartog A , Hautvast J. The consequences of seasonal food insecurity for individual food-consumption patterns in north-western Benin . Food Nutr Bull . 1995 ;
59. Iqbal A , Khalil IA , Ateeq N , Sayyar Khan M. Nutritional quality of important food legumes . Food Chem . 2006 ; 97 : 331 ± 335 . https://doi.org/10.1016/j.foodchem. 2005 . 05 .011
60. Ade-Omowaye BIO , Tucker GA , Smetanska I. Nutritional potential of nine underexploited legumes in Southwest Nigeria . Int Food Res J . 2015 ; 22 : 798 ± 806 .
61. Swindale A , Bilinsky P. Household Dietary Diversity Score (HDDS) for measurement of household food access: Indicator guide . Version 2. Food and Nutrition Technical Assistance (FANTA) . 2006 .
62. Rothman KJ , Greenland S , Lash TL. Modern-Epidemiology. Third. Lippincott Williams & Wilkins; 2008 .
63. Odusina O , Afolami CA , Momoh S. Seasonal Assessment of Household Food Access in Ibadan Metropolis, Nigeria . Discourse J Agric Food Sci . 2014 ; 2 : 53 ± 58 .
64. Tarasuk VS , Beaton GH . Women's dietary intakes in the context of household food insecurity . J Nutr . 1999 ; 129 : 672 ± 679 . PMID: 10082773
65. Obayelu AE . Households ` food security status and its determinants in the North-Central Nigeria . Food Econ . Taylor & Francis; 2012 ; 9 : 241 ± 256 . https://doi.org/10.1080/2164828X. 2013 .845559
66. Willet W. Nutritional Epidemiology . 3rd edition. Monographs in Epidemiology and Biostatstics. Third . Oxford University Press, USA; 2012 .
67. Martin-Prevel Y , Becquey E , Arimond M. Food group diversity indicators derived from qualitative listbased questionnaire misreported some foods compared to same indicators derived from quantitative 24-hour recall in urban Burkina Faso . J Nutr . 2010 ; 140 : 2086S± 93S . https://doi.org/10.3945/jn.110. 123380 PMID: 20881076
68. Gibson RS . Principles of Nutritional Assessment . Second. Oxford: Oxford University Press, USA; 2005 .