Introduction

Many low- and middle-income countries (LMICs) are going through a nutritional transition, with diets shifting toward more added sugars, unhealthy fats, salt, and refined carbohydrates [1]. As a result, children in both LMICs and industrialized countries consume more unhealthy foods and sugary drinks [2, 3]. Sugar-sweetened beverages are related to an increased risk of being overweight and obese due to excessive energy intake [4, 5]. Currently, a high intake of unhealthy foods throughout infancy and early childhood has been related to overweight and obesity [6,7,8,9,10].

Globally, 40 (5.9%) million children under the age of five are overweight or obese, with the majority of these children living in low- and middle-income countries [11]. Central Asia (14.9%) and the Middle East and North Africa (11.2%) have the highest rates of overweight [12]. According to the Ethiopia Demographic Health Survey (EDHS), the number of overweight children increased from 1 to 2% between 2016 and 2019 [13, 14]. Overweight and obesity rank as the fifth leading risk factor for mortality, and morbidity is one of the major public health problems, causing 45% of all deaths among children aged 0–59 months [15].

Childhood obesity has consequences for adulthood obesity, which can lead to psychosocial problems as well as cardiovascular risk factors such as high blood pressure, high cholesterol, diabetes, and sleep disordered breathing [16,17,18]. In addition, low self-esteem increases the likelihood of being bullied, as does poor school attendance and achievement, a low employment rate, and a lower-paying job in adulthood [19, 20]. Adult obesity also costs the world $ 2.0 trillion per year [21].

Reports show that food marketing environments [22], maternal occupation [23], the child’s first appointment at the primary healthcare facility [23], lower income households [23], TV food advertising [24], urban residence [25], the age of the child [25], and being enrolled in cash transfer programs [23] are the factors for increasing the consumption of unhealthy food.

Ethiopia has various initiatives and programs in place to minimize all forms of malnutrition, including overweight and obesity, as part of its national development plan since children under the age of two have been identified as being part of the first key window of opportunity [26]. However, the prevalence of obesity and overweight in Ethiopia ranges from 1.7 to 3.6% [27]. Furthermore, less attention has been paid to the increased accessibility of unhealthy foods, suboptimal feeding practices, and the rising trend in overweight/obesity. Regular monitoring of infant and child feeding practices helps to provide baseline evidence as well as monitor and evaluate current initiatives. However, little is known, including in the research area, regarding the prevalence of unhealthy food consumption among children aged 6 to 23 months in Ethiopia. As a result, the purpose of this study was to determine the prevalence and associated factors of unhealthy food consumption in 6 to 23 month old children in Gondar City, Northwest Ethiopia.

Methods

Study design, period and area

A community-based cross-sectional study was conducted from June 30 to July 21, 2022, in Gondar City, Northwest Ethiopia. Gondar City is the capital city of Gondar city administration and the central Gondar zone, which has six sub-cities. It is 727 km from Addis Ababa, the Federal Democratic Republic of Ethiopia’s capital, and 120 km from Bahir Dar, the capital of the Amhara National Regional State. The city has 3 hospitals, 8 health centers, and 14 health posts. The city has a total population of 454,445 people, including 22,945 children under the age of two.

Source population and study population

All infants and young children aged between 6 and 23 months, with their mothers or caregivers, living in Gondar city were considered the source population, whereas all infants and young children aged between 6 and 23 months, with their mothers or caregivers, in selected Kebeles living in Gondar city were considered the study population.

Sample size and sampling procedure

Based on the following assumptions, a sample size was calculated by using the single population formula: Because there have been no previous studies, a 95% CI was calculated using 50% of the previous prevalence, a 5% margin of error, and a 10% non-response rate. Finally, a total of 845 samples were obtained with 2% design effects. To identify study participants, a simple random sampling procedure was used. A list of infants and children was found in health posts. First, two sub-cities were selected from the total number of sub-cities in Gondar by using the lottery method. Second, six Kebeles were selected from each of the selected sub-cities by using the lottery method. Third, after having the number of all 6–23-month-old children from the selected kebeles, corresponding to the month of the 6-23-month-old children, samples were allocated proportionally to each kebele. Finally, a total of 845 participants were taken from each kebele through a simple random sampling technique.

Data collection procedures and tools

The data collection instruments were structured questionnaires that were pretested. The questionnaire was adapted from the World Health Organization (WHO) Infant and Young Child feeding (IYCF) indicators and other literature with some modifications [28,29,30,31,32]. The questionnaire was prepared originally in English and translated from Amharic back to English. It contains different independent variables, like the socio-demographic characteristics of parents, infants, and young children; household-related factors; and maternal health services related factors. The single 24-hour recall method was used to measure infants’ and children’s food consumption. There were four BSc nurse data collectors and one public health nutritionist supervisor assigned.

Data quality control

Two-day training was given on the contents of the questionnaire, interviewing technique, the purpose of the study, and how the principal investigator approaches and maintains the respondents’ confidentiality. A pre-test was conducted with 5% of the total sample size. The quality of the data was checked every day after data collection, and any errors were corrected before the following data gathering measures. Prior to analysis, the data was cleaned up and cross-checked. The principal investigator and the supervisor closely monitored the process throughout the data collection period and made any necessary corrections.

Variable measurnemts

To determine each child’s unhealthy food consumption (UFC), the mother was asked to list all foods consumed by the child in the 24 h preceding the survey, such as Juices, soda, coffee or tea with sugar, candies, chocolate, cakes, sweet biscuits, ice cream, potato chips and instant noodle. Children who got at least one food from the above lists were classified as meeting the unhealthy food consumption; otherwise, they were considered to be getting healthy food consumption [32].

A mother with less than two-year-old children who read a newspaper or magazine, listened to the radio, or watched television at least once per week is considered to have satisfactory media exposure; children who did so less than once per week are considered to have unsatisfactory media exposure.

Postnatal care (PNC) is defined as care provided to the mother and her newborn baby immediately following placental birth and for the first 42 days of life [33].

GMP utilization was measured by taking a child’s GMP and having them participate in GMP at least once in the previous three months.

The household wealth index was determined using principal component analysis (PCA) by considering household assets, such as quantity of cereal products, house, livestock, and agricultural land ownership. First, variables were coded between 0 and 1. Then, the variables were entered and analyzed using PCA, and those variables having a communality value of greater than 0.5 were used to produce factor scores. Finally, the factor scores were summed and ranked into poor, medium, and rich wealth index [34].

Data processing and analysis

Cleaned and coded data was entered into EPI Data Version 3.1 and exported to STATA Version 14 for further analysis. Descriptive and summary statistics were presented in the form of text, tables, and graphs. Both frequency and tables are used to summarize descriptive statistics. Variables with a p-value less than 0.2 in the bivariable analysis were fitted into the multivariable logistic regression analysis. A multivariable logistic regression analysis was employed to identify the factors associated with unhealthy food consumption. An adjusted odds ratio (AOR) with a 95% confidence interval was used to show the strength of the association, while a P-value of 0.05 was used to declare the significance of the association.

Results

Socio-demographic characteristics

A total of 811 mother-child pairs with children aged 6–23 months were enrolled in the study, with a response rate of 96%. The mean age of the respondents was 32.33 years, with a standard deviation of 6.24. More than one-third of the mothers, 277 (34.1%), were between the ages of 25 and 30. More than three-quarters of the respondents, 631 (77.8%) and 608 (75%), were married and Orthodox Christians, respectively. More than one third, 283 (34.9%), of the mothers were housewives. Less than a quarter, 182 (22.4%), and 160 (19.7%) of the mothers and fathers, were able to read and write their educational status, respectively. Housewives made up less than half of the 349 mothers (43%) (Table 1).

Table 1 Parental level characteristics of children aged 6–23 months in Gondar City, Northwest Ethiopia, 2022
Full size table

Child, household, and community-related characteristics

More than one-third, 297 (36.6%), of the children were within the age range of 12 to 17 months. The majority of study participants, 685 (84.5%), lived in urban Kebeles. More than half, 485 (59.8%) of respondents, had fewer than or equal to four family members. Most 724 (89.3%) of the participants did not have a garden at home. More than two-thirds 620 (76.40%) of the study participants, received advice on child feeding practices from health professionals. More than two thirds, 624 (76.94%) of the participants, began providing their children with complementary feeding, and 604 (74.5%) of the respondents had satisfactory media exposure. The vast majority of children, 730 (90%), are currently breastfed (Table 2).

Table 2 Child, household, and community characteristics of children aged 6–23 months in Gondar City, Northwest, Ethiopia, 2022
Full size table

Child, and health care level related characteristics

Nearly two thirds, 494 (60.9%) of the respondents’ mothers, had four and above ANC visits during their last pregnancies. Almost three-quarters 579(71.4%) of the mothers received information on infant feeding during any of their antenatal care (ANC) visits. The majority, 720 (88.8%) of the study participants, had PNC visits during their last pregnancies, and nearly two-thirds of 500 (61.7%) of the children had growth monitoring and promotion (GMP) in the last three months. Regarding the birth place, the majority 773, (95.3%) of children were from health facilities (Table 3).

Table 3 Child and health care level related factors among 6-23months of child in Gondar City, Northwest Ethiopia, 2022
Full size table

Prevalence of unhealthy food consumption

The overall prevalence of unhealthy food consumption among children aged 6–23 months was 63.7% (95% (CI: 60.4, 67.2%).

Factors associated with unhealthy food consumption

The multivariable logistic regression revealed that educational status of the mother, age of the child; urban residence, family size, and GMP service were all significantly associated with unhealthy food consumption (Table 4).

Table 4 Multivariable logistic regression of variables for unhealthy food consumption among children 6-23months of age in Gondar City, Northwest Ethiopia, 2022
Full size table

Mothers unable to read and write were 1.89 times [AOR = 1.89, 95%CI = 1.05, 3.69] more likely to consume unhealthy food as compared to those with a diploma and above.

Children aged 18–23 months were 47% less likely [AOR = 0.53; 95% CI = 0.34; 0.74] to consume unhealthy food than those aged 6–11 months.

Children living in families larger than four were 1.22 times more likely to consume unhealthy food than those living in families smaller than or equal to four [AOR = 1.22, 95% CI = 1.07, 2.78].

Moreover, mothers who lived in urban areas were 4.55 times more likely [AOR = 4.55, 95% CI = 3.61, 7.78] to provide unhealthy food as compared to mothers who lived in rural areas.

Mothers who had not received GMP services were 2.07 [AOR = 2.07, 95% CI = 1.48, 3.18] times more likely to provide unhealthy foods as compared with mothers who had received GMP services.

Discussion

Unhealthy food consumption among children aged 6–23 months in Gondar city was 63.7%. This finding is consistent with a study done in Iran (66.8%) [35]. Whereas, this report is lower than a study done in Nepal (74.1%) [36]. The possible reason might be due to different sociodemogrphic characteristics, cultural variation, and study design. However, this finding is higher than in Tanzania (23.10%) [37], Dakar (58.7%) [29], and Peru (19.30%) [38]. One possible explanation is that the Peru study was conducted during the COVID-19 pandemic, which limited market access to unhealthy foods [39].

Mothers unable to read and write were 1.89 times more likely to consume unhealthy food as compared to those with a diploma and above. The possible explanation might be that women who are unable to read or write are unconcerned about the short- and long-term consequences of unhealthy food consumption [40]. In addition, mothers who are unable to read or write will have poor knowledge of nutrients, limited access to high-quality food sources, lower incomes, and are less likely to buy recommended health foods, preferring instead to buy sugary beverages or sweet baked foods [41, 42].

Children living in families larger than four were 1.22 times more likely to consume unhealthy food than those living in families smaller than or equal to four. This could be due to the fact that mothers with larger families take longer to prepare foods and, as a result, purchase unhealthy items for their family. Furthermore, intra-household food distribution may boost the desire to consume unhealthy foods [43]; unhealthy foods are less expensive and easier to prepare than healthy foods [28].

Moreover, mothers who lived in urban areas were 4.55 times more likely to provide unhealthy food as compared to mothers who lived in rural areas. The finding is supported by the global food policy report [44]. The possible reason might be due to the fact that mothers living in urban residences may have the chance to access and purchase different kinds of unhealthy food from small shops, supermarkets, and street food vendors to provide unhealthy food for their children. Mothers who are living residences may be affected by food insecurity [45], shifting to purchasing less costly food from the market [46]. The other possible reason might be the urban poor population, for whom the most easily available and affordable diets are often unhealthy foods [47].

Children aged 18–23 months were 47% less likely to consume unhealthy food than those aged 6–11 months. Our finding is contradicted by studies done in Africa, Asia [29], and Nepal [48]. Possible reasons include child preference and a strong demand for sweet and convenient foods to begin complementary feeding at a young age. In addition, 18- to 23-month-old children have a higher probability of eating a family diet, which decreases unhealthy food consumption [49].

Mothers who hadn’t had GMP visits were 2.07 times more likely to provide unhealthy foods as compared to mothers who had GMP visits. The possible reason might be due to the fact that GMP focuses on empowering mothers to know about and become competent in appropriate child care and feeding practices through individual and group child feeding practices [50]. GMP is one of the techniques used to develop mothers’ and caregivers’ awareness, knowledge, and skill in preparing a diverse diet at home through community conversion and porridge preparation rather than providing unhealthy food for their children [51].

Strength and limitation of the study

The strength of this study was that it was the first of its kind in the study area targeting unhealthy food consumption among children aged 6–23 months. As a limitation, the data collection was done based on past feeding practices, which might be socially desirable. Since the data collection was done in a single 24-hour recall, the finding did not indicate the usual dietary habit of an individual child.

Conclusion

In Gondar City, nearly two thirds of infants and children received unhealthy food. Maternal education, urban residence, GMP service, child age, and family size were all significant predictors of unhealthy food consumption. Thus, improving the uptake of GMP services and family planning services is critical to reducing unhealthy food consumption.