Objective To assess the association between socioeconomic status (SES) and dietary sodium intake, and to identify if the major dietary sources of sodium differ by socioeconomic group in a nationally representative sample of Australian children.
Design Cross-sectional survey.
Setting 2007 Australian National Children's Nutrition and Physical Activity Survey.
Participants A total of 4487 children aged 2–16 years completed all components of the survey.
Primary and secondary outcome measures Sodium intake was determined via one 24 h dietary recall. The population proportion formula was used to identify the major sources of dietary salt. SES was defined by the level of education attained by the primary carer. In addition, parental income was used as a secondary indicator of SES.
Results Dietary sodium intake of children of low SES background was 2576 (SEM 42) mg/day (salt equivalent 6.6 (0.1) g/day), which was greater than that of children of high SES background 2370 (35) mg/day (salt 6.1 (0.1) g/day; p<0.001). After adjustment for age, gender, energy intake and body mass index, low SES children consumed 195 mg/day (salt 0.5 g/day) more sodium than high SES children (p<0.001). Low SES children had a greater intake of sodium from processed meat, gravies/sauces, pastries, breakfast cereals, potatoes and potato snacks (all p<0.05).
Conclusions Australian children from a low SES background have on average a 9% greater intake of sodium from food sources compared with those from a high SES background. Understanding the socioeconomic patterning of salt intake during childhood should be considered in interventions to reduce cardiovascular disease.
- Nutrition & Dietetics
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-commercial License, which permits use, distribution, and reproduction in any medium, provided the original work is properly cited, the use is non commercial and is otherwise in compliance with the license. See: http://creativecommons.org/licenses/by-nc/2.0/ and http://creativecommons.org/licenses/by-nc/2.0/legalcode.
Statistics from Altmetric.com
To assess the association between socioeconomic status (SES) and dietary sodium intake in Australian children and adolescents.
To determine if the major dietary sources of sodium differ by socioeconomic group.
In Australian children, SES is inversely associated with dietary sodium intake.
Children of low socioeconomic background consumed more sodium from convenience style foods including pies/sausage rolls; savoury sauces, fried prepared potato; processed meat and potato crisps.
Strengths and limitations of this study
These results are based on a large nationally representative sample of Australian children and adolescents.
Sodium intake was determined via a 24 h dietary recall and therefore does not capture the amount of sodium derived from salt used at the table or during cooking.
The socioeconomic disparity of sodium intake reported in this study is attributable to differences in sodium intake from food sources only. Further research is required to understand how SES impacts on raising sodium intake.
As in adults,1 dietary sodium intake is positively associated with blood pressure in children.2 ,3 Comparable to other developed nations,4 the dietary sodium intake of Australian children is high and exceeds dietary recommendations.5 ,6 Given that blood pressure follows a tracking pattern over the life course,7 ,8 it is very likely that high sodium consumption during childhood increases future risk of adult hypertension and subsequent cardiovascular disease (CVD). Increased CVD risk is also observed with low socioeconomic status (SES), 9 ,10 potentially due in part to differences in dietary intake. Furthermore, prolonged inequalities of SES across the life course are likely to accumulate to overall greater CVD risk,11 ,12 A number of studies in adults13–15 and in children and adolescents16–20 have identified SES as a determinant of diet quality. For instance, evidence from cross-sectional studies in children and adolescents have reported a positive association between SES and fruit and vegetable intake,17 ,18 ,21 and conversely, lower levels of SES have been associated with poor dietary outcomes, including a greater intake of high fat foods,20 fast foods and soft drinks.19 Studies examining the association between SES and sodium intake are scarce and inconsistent. One study in British adults found that low SES was associated with a higher intake of sodium,22 whereas in US adults, there was no association between SES and sodium intake.23 The aim of this study was to examine the association between SES and dietary sodium intake and the food sources of sodium in a nationally representative sample of Australian children aged 2–16 years.
The 2007 Australian Children's Nutrition and Physical Activity Survey (CNPAS) was a cross-sectional survey designed to collect demographic, dietary, anthropometric and physical activity data from a nationally representative sample of children aged 2–16 years. The full details of the sampling methodology can be found elsewhere.24 Briefly, participants were recruited using a multistage quota sampling framework. The initial target quota was 1000 participants for each of the following age groups: 2–3, 4–8, 9–13 and 14–16 years (50% boys and 50% girls), to which a 400 booster sample was later provided by the state of South Australia. The primary sampling unit was postcode and clusters of postcodes were randomly selected as stratified by state/territory and by capital city statistical division or rest of state/territory. Randomly selected clusters of postcodes ensured an equal number of participants in each age group, from each of the metro and non-metro areas within each state. Within selected postcodes, Random Digit Dialling was used to invite eligible households, that is, those with children aged 2–16 years, to participate in the study. Only one child from each household could participate in the study. The response rate of eligible children was 40%. Owing to the non-proportionate nature of the sampling framework, each participant was assigned a population weighting which weighted for age, gender and region. The study was approved by the National Health and Medical Research Council registered Ethics Committees of the Commonwealth Scientific and Industrial Research Organisation and the University of South Australia. All participants or, where the child was aged <14 years, the primary carer provided written consent.24
Demographic and food intake data were collected during a face-to-face computer-assisted personal interview completed between February 2007 and August 2007. A three-pass 24 h dietary recall was used to determine all food and beverages consumed from midnight to midnight on the day prior to the interview.24 The three-pass method includes the following stages: (1) providing a quick list of all foods and beverages, (2) a series of probe questions relevant to each quick list item to gather more detailed information on the time and place of consumption, any additions to the food item, portion size and brand name and (3) finally, a recall review to validate information and make any necessary adjustments. Portion sizes were estimated using a validated food model booklet and standard household measures. To minimise error after data collection, all interviews were reviewed by study dieticians to assess for unrealistic portion sizes, inadequate detail and typing errors. The primary carer of participants aged 9 years and under provided information on dietary intake.24
Sodium intake was calculated using the Australian nutrient composition database AUSNUT2007, specifically developed by the Food Standards Australia and New Zealand for the CNPAS.25 The food coding system used in this database has previously been described.26 Daily sodium (mg) intake was converted to the salt equivalent (g) using the conversion 1 gram of sodium chloride (salt)=390 mg sodium. The reported salt intake did not include the salt added at the table or during cooking.
Indicator of socioeconomic status
Consistent with the other dietary studies in children and adolescents, we have used the level of education attained by the primary carer and household income as markers of SES.27 ,28 The highest level of education attained by the primary carer was used to define SES. Based on this, participants were grouped into one of three categories of SES: (1) high: includes those with a university/tertiary qualification, (2) mid: includes those with an advanced diploma, diploma or certificate III/IV or trade certificate and (3) low: includes those with some or no level of high school education. Parental income was used as a secondary indicator of SES. Reported parental income before tax was grouped into four categories (1) AUD$ 0 to $31 999, (2) $31 200 to $51 999, (3) $52 000 to $103 999 and (4) ≥$104 000. Body weight and height were measured using standardised protocols.29 Body mass index (BMI) was calculated as body weight (kg) divided by the square of body height (m2). Participants were grouped into weight categories (very underweight, underweight, healthy weight, overweight, obese) using the International Obesity Task Force BMI reference cut-offs for children.30 ,31
Statistical analyses were completed using STATA/SE V.11 (StataCorp, College Station, Texas, USA) and PASW Statistics V.17.0 (PASW Inc, Chicago, Illinois, USA). A p value of <0.05 was considered significant. All analyses accounted for the complex survey design using the STATA svy command, specifying the strata variable (region), cluster variable (postcode) and population weighting (age, gender and region). Descriptive statistics are presented as mean (SD) or n (% weighted). Pearson correlation coefficient was used to assess the association between sodium intake, energy intake and BMI. To assess the association between SES, as defined by primary carer education level, and sodium intake, multiple regression analysis was used with adjustment for age, gender, energy intake and BMI. To further control for the effects of age, the analysis was repeated stratified by the age group (ie, 2–3; 4–8; 9–13; 14–16 years). These age categories are consistent with those used in Australian dietary guidelines6. As income level is sometimes used as a marker of SES, 13 the association between parental income and sodium intake was also examined, with adjustment for age, gender, energy intake and BMI. The regression coefficient (β) with 95% CI, corresponding p values and the coefficient of determination (R2) are presented. In a previous analysis26 which included the same study population, we used the population proportion formula32 to calculate the contribution of sodium from submajor food group categories, as defined in the CNPAS food group coding system.24 The population proportion formula32 is outlined below:
For the present study, we have utilised this list, which identifies the main sources of dietary sodium, to determine if sodium intake from the food group differs between low and high SES categories, based on primary carer education level. To do this, we calculated the mean sodium intake from each submajor food group by SES category, and compared the mean sodium of low with high SES, using an independent t test.
Basic characteristics of the 4487 participants are listed in table 1. As defined by parental education status, the proportion of children from low, medium and high SES backgrounds was relatively evenly distributed. Over two-thirds of children fell within the two highest income bands. There was a significant positive correlation between sodium intake and energy intake (r=0.69, p<0.001) and sodium intake and BMI (r=0.22, p<0.001). Average daily sodium intake differed by SES (figure 1, p<0.01). Regression analysis indicated that low SES was associated with a 195 mg/day (salt 0.5 g/day) greater intake of sodium. The association between SES and sodium intake remained after adjustment for age, gender, energy intake and BMI (table 2). When stratified by the age group, the association between sodium intake and SES remained significant between the ages of 4–13 years (table 2); however, there was no association between sodium intake and SES in 2-year-olds to 3-year-olds or in 14-year-olds to 16-year-olds (data not shown). There was no association between sodium intake and parental income (data not shown); however, only 28% of children fell within the two lowest income bands (table 1). Table 3 lists those submajor food groups which contributed >1% to the groups’ total daily sodium intake. Combined, these 23 food groups accounted for 84.5% of the total daily sodium intake. Regular breads and bread rolls contributed the most sodium (13.4%). Moderate sources of sodium, contributing more than 4% of the total sodium intake, included mixed dishes where cereal is the major ingredient (eg, pizza, hamburger, sandwich, savoury rice and noodle-based dishes), processed meat, gravies and savoury sauces, pastries, cheese and breakfast cereals and bars. Compared with children of high SES, children of low SES had a significantly greater intake of sodium from processed meat, gravies and savoury sauce, pastries, breakfast cereals and bars, potatoes and potato snacks (eg, potato crisps). The percentage difference in sodium intake in each of these categories was 46%, 31%, 24%, 16%, 39% and 46%, respectively (table 3). Conversely, children of high SES background had a significantly greater intake of sodium from the food group containing cakes, buns, muffins, scones, cake-type desserts, and from the food group described as batter-based products (eg, pancakes and picklets). The percentage difference in sodium intake in each of these categories was 16% and 32%, respectively (table 3).
In a nationally representative sample of Australian children aged 2–16 years, we found that children of low SES background consumed 9% more dietary sodium from food sources than those of high SES background. The inverse association between sodium intake and SES was primarily driven by the association in children aged 4–13 years, particularly after adjustment for the important covariates age, gender, energy intake and BMI. In adult studies, low SES has been associated with more frequent consumption of high salt foods, such as soup, sauces, ready-to-eat meals, savoury seasonings, sausages and potato.33 ,34 Given the parental control over children's food choices during these years, it is likely that SES disparities in adult food choices relating to high-salt foods may filter down into children's eating practices. We found no association between SES and sodium intake in 2-year-olds to 3-year-olds and 14-year-olds to 16-year-olds. Although some evidence indicates that SES disparities in dietary patterns may be present during infancy,35 it is possible that such early differences are not seen in dietary patterns with the restricted range of food types. In the case of adolescents, as autonomy over food choices increases, other factors, such as peer influence, taste and eating away from the home,36 may become more prominent determinants of dietary intake.
Using US National Health and Nutrition Examination Survey data, Mazur et al37 explored the association of SES, as indicated by the head of household education status and household income, on sodium intake in Hispanic children aged 4–16 years. Interestingly, in this study, lower levels of education were associated with lower sodium intake.37 This is in contrast to our own findings as well as past studies, which generally link lower SES to overall poorer dietary outcomes.13 We found no association between sodium intake and level of income; however, low-income bands were under-represented. This is in contrast to the findings in Hispanic children, where low household income was associated with a greater intake of dietary sodium.37 In a New Zealand food survey, low-cost ‘home brand’-labelled food products were found to contain greater quantities of sodium than the more expensive branded food products.38 The impact of income on sodium intake in Australian children remains unclear and further research is required.
Previous studies in children have reported socioeconomic differences in the consumption of certain food groups.39 ,40 For example, in European children of low SES background, greater intake of starchy foods, meat products, savoury snacks such as hamburgers, sugar and confectionery, pizza, desserts and soft drinks have been reported.39 ,40 In the present study, those food groups which were found to contribute more sodium to the diets of low SES children tended to include convenience style foods (ie, pies/sausage rolls; savoury sauce and casserole base sauces; fried prepared potato; processed meat and potato snacks). Comparably, children of high SES background consumed greater amounts of sodium from cake-type and baked-type products. However, a significant amount of sodium in baked products can be in the form of sodium bicarbonate rather than sodium chloride. Sodium bicarbonate, unlike sodium chloride, has not been directly associated with adverse blood pressure outcomes.41
With reference to sodium intake data by age group5 and in comparison to the recommended daily upper limit of sodium,6 it is evident that Australian children of all ages across all SES backgrounds are consuming too much dietary sodium. However, for the first time, our findings indicate that socioeconomic disparities exist in sodium intake in Australian children aged 9–13 years. To reduce sodium intake in children, a comprehensive approach is required: first, targeting food policy to encourage product reformulation of lower sodium food products across all price ranges within the food supply. Second, consumer education and awareness campaigns that encourage food choices which are based on fresh products with minimal processing; this may require strategies that equip parents with enhanced food preparation skills and knowledge of the ‘hidden’ salt added to many commonly eaten processed foods. Furthermore, it is apparent that these strategies need to reach lower SES groups.
The major strengths of this study include the use of a large nationally representative sample of Australian children, with a comprehensive and standardised collection of dietary intake. Limitations of the study include the use of a 24-h dietary recall to assess sodium intake. First, this method fails to capture the amount of salt derived from salt added at the table and during cooking and therefore is likely to underestimate the true value of salt intake.42 The majority (77%) of dietary sodium consumed is from salt added to processed foods, while a smaller amount (11%) has been found to be derived from salt added at the table and during cooking.43 In the present study, the higher intake of sodium reported in children from low SES background is attributable to differences in sodium intake from food sources only. In a previous analysis of these data, we found that children from low SES background (33%) were more likely to report adding salt at the table than children from high SES (25%).26 Thus, it is likely that children of low SES background are consuming greater amounts of total daily sodium than reported in the present analysis. Second, the assessment of sodium intake is limited by the quality of food composition databases, which may not capture the variation in sodium content of different brand products within each food group. 42 ,44
In summary, the findings of higher salt intake from food sources in children of lower SES background, within a nationally representative sample, provide focus for concern regarding salt-related disease across the life course. This socioeconomic patterning of salt intake may, in turn, influence the SES disparity seen in hypertension and cardiovascular risk in adulthood. To reduce the socioeconomic inequalities in health, interventions need to begin early in life and should include product reformulation of lower sodium food products across all price ranges, as well as consumer education and awareness campaigns which reach low SES groups.
Contributors The author's responsibilities were as follows—CAG, KJC, LJR and CAN designed the research; CAG performed the statistical analysis and wrote the manuscript and is also the guarantor of the paper; LJR, KJC and CAN helped with data interpretation, revision of the manuscript and provided significant consultation. All authors have read and approved the final manuscript.
Funding This work was supported by a postgraduate scholarship from the Heart Foundation of Australia PP 08M 4074.
Competing interests All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: Carley Grimes had financial support in the form of a postgraduate scholarship from the Heart Foundation, Australia for the submitted work and had no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years, no other relationships or activities that could appear to have influenced the submitted work. Karen Campbell had no support from any organisation for the submitted work; no other relationships or activities that could appear to have influenced the submitted work. Caryl Nowson has received research funds from Meat & Livestock Australia; National Health and Medical Research Council, Wicking Foundation, National Heart Foundation, Australia, Helen MacPerhson Smith Trust and Red Cross Blood Bank. These payments are unrelated to the submitted work. Lynn Riddell has received research funds from Meat & Livestock Australia. These payments are unrelated to the submitted work.
Patient consent Obtained.
Ethics approval National Health and Medical Research Council registered Ethics Committees of Commonwealth Scientific and Industrial Research Organisation and the University of South Australia.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No additional data are available.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.