Introduction

Periods of childhood growth are sensitive, yet there are critical windows that are widely influenced by nutritional status. They require more energy and nutrients such as iron and zinc to support adequate growth [1]. It has been previously reported in Iranian children aged 2–5 years old that the prevalence of stunted childhood growth was 7.5%, rates of underweight and overweight 6.9% and 17.8% in boys and 4.7% and 11.7% in girls, respectively [2]. It is well understood that weight and height development at an early age can critically impact health quality in later life. While childhood overweight or obesity has been associated with comorbidities such as asthma, diabetes, metabolic disease, hypertension and cardiovascular diseases (CVD) later in life [3], children and adolescents who fall below recommended weight categories are associated with severe anemia and heart failure [4]. The rising demand for adequate food in childhood and adolescence requires adequate income and food availability and accessibility. The imbalance between cost of living and income has resulted in rapidly increasing levels of food insecurity among households all over the world specifically in developing countries [5]. Lack of money to preparing food is a blockade for having a healthy weight [6]. Previous research reported that household food insecurity among Iranian adolescents was 41% and 67% in children [7, 8].

Food insecurity is the unavailability of nutritious, sufficient and quality food and has been linked to both malnutrition and obesity [9, 10]. Ultimately, food insecurity can lead to negative health outcomes such as nutrient deficiency, anxiety, depression, decreased physical function, academic performance and chronic disease [7, 11]. Childhood nutritional status such as underweight, overweight or stunted growth is largely impacted by food insecurity, it has an influence on food quantity, quality, accessibility and availability [12]. As such, previous literature indicates food insecurity is widely associated with dietary patterns. However, these results are inconsistent in different study populations [13]. A healthy diet is an essential factor for not only physical development but cognitive growth [7]. Dietary acid load (DAL) is an index that determines the balance of acid-inducing foods such as meat and eggs and base-inducing foods such as fruits and vegetables [14]. DAL can be estimated with three commonly used scores: net endogenous acid production (NEAP) that is calculated as the ratio of ingested protein and potassium, potential renal acid load (PRAL) calculated by calcium, phosphorus, magnesium, potassium and protein, or using the ratio of animal protein-to-potassium [15, 16]. Acid–base imbalance is recognized as a risk factor for metabolic disorder, previous research among German children has demonstrated a higher dietary acid load was associated with higher blood pressure and loss of bone mineral contents [17].

The dietary energy density (DED) is the amount of available energy per unit of weight in the diet (kJ/g). DED is a marker of diet quality [18] which has been previously demonstrated to be related to overeating, overweight, weight gain, childhood obesity and changes in body composition [19, 20]. In contrast, lower DED is associated with better nutritional quality and a higher intake of fiber [21]. A study conducted in American children aged 2–8 years old found children with a lower DED similarly had a lower energy intake of fat and consumed more fruit and vegetables [22]. Moreover, energy-dense foods are less expensive and more available to buy for lower socioeconomic households [23, 24].

To our best knowledge, there is no study that has evaluated the association of food insecurity and dietary acid load with anthropometric status of children. The impact of food insecurity and food quality indicators on the growth of children highlights the importance of the present study aimed to explore the association of food insecurity and dietary acid load with anthropometric status of children.

Methods

This cross-sectional study observed 788 six-year-old girls referred to health centers affiliated to Tehran University of Medical Sciences who referred to health care centers affiliated to Tehran University of Medical Sciences for vaccination in south of Tehran from October 2017 to March 2018. Multi-stage cluster sampling was conducted. Inclusion criteria included Iranian children with obtained parental consent while children with diagnosed cardiovascular, liver and kidney diseases and cancer were excluded. General information, the food frequency questionnaire, and the United States Department of Agriculture (USDA) 18-item food security questionnaire were filled by the mother or maternal figures of participants.

Food security score was calculated by the total of responses score of USDA 18-item food security questionnaire previously validated [25]. Each item was rated from “0” to “2” points. Families which receive a total score of 0–2 are characterized as food secure, while a score of “3” to “7” and “8” to “12” indicate food insecurity without hunger and food insecurity with moderate hunger, respectively, and a score of “13” or higher indicates food insecurity with severe hunger. Finally, food security status has been divided into food secure and food insecure individuals.

Children supplement consumption was assessed in the general questionnaire, and their physical activity score was calculated based on metabolic equivalent by a reported weekly timetable [26].

Socioeconomic status (SES) was evaluated using a comprehensive questionnaire which included several items such as parent’s occupation and education, number of children under 18 years old, number of family members, income level, health care insurance, food assistance, being the owner or tenant of a house, neighbourhood, vehicle ownership, number of cars and number of rooms in the house.

Anthropometric measurements including height, weight, and body mass index (BMI) were evaluated. Weight was measured by a digital scale (0.1 kg accuracy) with minimum clothing and no shoes. Height was measured using a fixed plastic meter to the wall in standard standing position. BMI was calculated by dividing the weight (kg) to the square of height (m). Parameters outlined by the World Health Organization (WHO) and Anthroplus software (version 1.0.4) were used to calculate the z-scores for BMI-for-age (BAZ), height‐for‐age (HAZ), and weight-for-age (WAZ). Therefore, BAZ ≥ 1 standard deviation (SD) was considered as overweight, BAZ ≥ 2 SD as obesity, − 3 ≤ BAZ ≤ − 1 as thin and BAZ ≤ − 2 SD was considered as severely thin. Children with HAZ ≤ − 2 SD and ≤ − 3 SD was defined as stunted and severely stunted, respectively. However, there were no participants in the present study characterized as stunted [27].

A validated food frequency questionnaire (FFQ 168 items) was used to evaluate participant’s dietary intake [28]. Frequency of food item consumption was considered by the day, week, month and year. USDA household consumption values were used for computing grams per day for each food item. NUTRITIONIST-IV software modified for Iranian foods (version 7.0; NSquared Computing, Salem, OR) was used to estimate nutrient and energy intake [29].

Dietary acid load calculated based on the PRAL method outlined below [30].

$$\begin{aligned} & {\text{PRAL }}\left( {{\text{mEq}}/{\text{day}}} \right) \, = \, \left( {{\text{protein }}\left[ {{\text{g}}/{\text{day}}} \right] \, \times \, 0.{49}} \right) \, + \, \left( {{\text{P }}\left[ {{\text{mg}}/{\text{day}}} \right] \, \times \, 0.0{37}} \right) \, {-} \, \left( {{\text{K }}\left[ {{\text{mg}}/{\text{day}}} \right] \, \times \, 0.0{21}} \right) \, \\ & \quad {-} \, \left( {{\text{Ca }}\left[ {{\text{mg}}/{\text{day}}} \right] \, \times \, 0.0{13}} \right) \, {-} \, \left( {{\text{mg }}\left[ {{\text{Mg}}/{\text{day}}} \right] \, \times \, 0.0{26}} \right). \\ \end{aligned}$$

Dietary energy density was calculated by dividing daily energy intake (kcal/day) into the total weight of food consumed (g/day) [31]. Only food intake was considered, drinks and beverages were excluded from analysis. Therefore, the effect of DED on body weight is related to only changes in food intake consumed, not drinks [32].

Normality of data was checked by histogram curve and Kolmogorov–Smirnov test. Food security status was reported as “food secure” and “food insecure”. Qualitative and quantitative variables across dietary scores were assessed using a Chi-square and independent sample t test and one-way ANOVA test, respectively. Food intake of children was classified into the median of dietary scores. Dietary intakes were compared using the analysis of covariance by adjusting the effect of energy intake among the median of dietary scores. To determine the relationship between food security status and anthropometric growth in children, binary logistic regression was used in the crude and adjusted model (Model 1). Physical activity, energy intake, SES, and supplement consumption were adjusted for in model 1. To determine the relationship between DAL and anthropometric growth in children, binary logistic regression was used in the crude and adjusted model. In the adjusted model, further adjusting was performed for food security status. SPSS software was used to perform analyses and p < 0.05 was considered as statistically significant.

Results

In the present study, 526 families were food secure. Mean weight of all children was 20.85 ± 2.35 kg. Table 1 shows the characteristics of participants across DAL and DED scores. There was no significant association between demographic and anthropometric variables and dietary scores. Table 2 presents the dietary intakes according to DAL and DED scores. Protein intake was positively associated with a high DAL score, while energy intake was positively associated with DED. Intake of energy, fat, carbohydrate, fiber, potassium, phosphorus, calcium, iron, folate, vitamin B12, mono- and poly-unsaturated fatty acids was inversely associated with DAL.

Table 1 Demographic and anthropometric characteristics of children according to dietary acid load and dietary energy density (n = 788)
Full size table
Table 2 Dietary intakes of children according to dietary acid load and dietary energy density (n = 788)
Full size table

The multivariable-adjusted odds ratio for anthropometric disorders and dietary scores across the food security status is indicated in Table 3. Our findings suggest food insecure participants are more likely to be classified as thin (OR 5.36; 95% CI 3.41–8.40) than overweight (OR 0.18; 95% CI 0.12–027) or obese (OR 0.28; 95% CI 0.08–0.98). There was no significant association between DED, PRAL, and food security status. Supplemental Table 1 presents the Z scores of anthropometric variables according to food security status which confirms the results of Table 3. Also, there was no significant association between the mid-arm circumference and food security status. Table 4 shows that there was no significant association between anthropometric measure and PRAL.

Table 3 Multivariable-adjusted odds ratio for anthropometric disorders and adherence to dietary scores according to food security status (n = 788)
Full size table
Table 4 Multivariable-adjusted odds ratio for anthropometric disorders according to dietary acid load (n = 788)
Full size table

Discussion

In this cross-sectional study, we examined the association between food insecurity, dietary acid load, dietary energy density and anthropometric indices among 788 6-year-old Iranian children. Our findings dictate that there was no correlation between DAL, DED and food insecurity. However, food insecure children were more likely to be classified as thin compared to children categorized as food secure. Finally, there was no association between DAL and anthropometric measures.

A study which assessed 300 healthy mother-infant pairs revealed no significant associations between DED quartiles among lactating mothers and infant growth between 2 and 4 months of age [33]. In contrast, a previous study demonstrated that DED was positively associated with BMI, fat mass index, body fat percentage and abdominal fat in 8–12-year-old children. It should be noted that Gomez-Burton et al. looked at both genders in addition to an older age category [34]. While previous evidence has suggested that DED is positively associated with cardiometabolic risk in children, our findings indicate a lack of association between these two variables. However, this may be indicative of a homogeneous group of children who were participated in our study sample. Moreover, overweight or obese children with different age (8–12 years old) can be another cause of diverse results which in this age, children have various dietary habits regarding to hormonal changes, puberty and independency feeling. Therefore, children in this age may eat fast foods, sugar sweetened beverage, snacks and other high energy dense foods which are related to overweight and obesity. Also, differences anthropometric evaluation should be considered. In the present study, weight and height were assessed, however, Gomez-Buurton et al. assessed the body composition [34, 35].

There is a limited body of research which identifies the association of DAL or DED with food security status among children. A cross-sectional study among 525 Iranian girls aged 14–18 years old demonstrated a positive association between dietary inflammatory index (DII) and food insecurity; DII is a dietary score which can be used to assess diet quality [36]. Further, a cross-sectional study which evaluated 218 children aged 10 years old using a healthy eating index concluded that children from insecure households reported reduced energy intake and reduced intake of whole fruits compared with children characterized as food secure [28].

The present study suggests that participants who were food insecure were more likely to be classified as thin and less likely to be classified as overweight or obese. Anthropometric measurements such as weight and height are reflective of nutritional status, and as such can be indicative of household social and economic status [37, 38]. Indeed, previous research determined that food insecurity was associated with low economic status [39]. Further, previous research has identified the relationship between food insecurity and poor health outcomes in children including emotional, behavioral, and academic performance, blood pressure, and increased hospitalization during early childhood [7, 40, 41]. A cross-sectional study in 150 Iranian households with 1–1.5-year-old infants revealed that infant height was significantly lower socioeconomic households [42]. Moreover, recently a study has revealed that adults who experience food insecurity report intentional dietary restraint. Also, this dietary restraint was associated with eating disorders in food insecure families which can effect on anthropometric indices by itself [43]. Also, one a cross-sectional study has indicated that food insecurity was associated with self-reported bulimia nervosa as well as greater binge-eating frequency [44]. Since food intake is an important factor in optimal growth, it is critical to develop proactive food policies and increase public awareness to the importance and accessibility of proper nutrition from a young age. Several approaches such as diet quality indices like as DAL, DED, and assessments of food insecurity can be used to evaluate the association between food intake and health outcomes. Despite this, Joe et al. established a very weak correlation between poor anthropometric measures and food intake suggesting that deficient anthropometric measures were not positively associated with food intake in Indian children. This effect was observed following statistical adjustments for socioeconomic and maternal health status [45]. South et al. indicated that 2.1% of food insecure children aged 8–17 years were underweight, while 44.2% of them were overweight or obese; food insecurity was further associated with increased blood pressure in participants [40].

The current study was conducted to assess the association between DAL, DED, food security and anthropometric measures in Iran, however several limitations should be addressed. This study included a heterogeneous sample of 6-year-old children which included thin, overweight and obese participants. A future study should be conducted in a more exclusionary population to delineate further if any associations occur. This study evaluated the physical activity level of participants, however future studies should assess body composition with more accurate methods such as Dual Energy X-ray Absorptiometry Scan (DXA) or BodPod. For the correct determination of food consumption, it was better if food consumption had been determined by a different method in addition to the food frequency questionnaire. Finally, confounding variables may have had an unknown impact on results, but future studies are needed.

Finding from our study revealed that there was no association between DAL with food insecurity. However, the likelihood of being classified as thin was higher in food insecure children. Overall, food policies and educational programs should be provided to prevent growth abnormalities in Iranian children. More well-designed studies such as cohort designs need to be performed in different countries and in both genders to confirm findings presented in this study.

What is already known on this subject?

No study has evaluated the association of food insecurity and dietary acid load with anthropometric status of children. The impact of food insecurity and food quality indicators on the growth of children highlights the importance of the present study aimed to explore the association of food insecurity and dietary acid load with anthropometric status of children.

What does this study add?

The findings from our study revealed that there was no association between DAL with food insecurity. In the present study, protein and energy intakes were higher in children with a higher dietary acid load (DAL) and dietary energy density (DED) score, respectively. Energy, fat, carbohydrate, fiber, potassium, phosphorus, calcium, iron, folate, vitamin B12, mono- and poly-unsaturated fatty acid intake was inversely related to DAL. Children characterized as food insecure were more likely to defined as thin (OR 5.36; 95% CI 3.41–8.40) than overweight (OR 0.18; 95% CI 0.12–027) and obese (OR 0.28; 95% CI 0.08–0.98, respectively). There was no significant association between DED, potential renal acid load (PRAL), and food security status. Moreover, there was no significant association between anthropometric measure and PRAL. Overall, food policies and educational programs should be provided to prevent growth abnormalities in Iranian children.