Background
Indigenous Australians experience premature mortality due to chronic disease at a highly disproportionate rate, and much earlier age, compared with non-Indigenous Australians [
1]. Risk of cardiovascular disease (CVD) mortality in Indigenous Australians is nearly twice that of non-Indigenous Australians [
2], and CVD is responsible for approximately 3 years of the life-expectancy gap experienced by this population [
1]. The high prevalence of chronic kidney disease (CKD) in the Indigenous Australian population is growing concern, particularly in very remote areas; nearly four in ten Indigenous Australians living in very remote Australia have indicators of CKD [
3]. Dietary improvement strategies are a priority for reducing chronic disease risk and improving health equity between Indigenous and non-Indigenous Australians.
The World Health Organisation (WHO) recommends reduction in dietary salt as a cost-effective strategy to reduce risk of chronic disease [
4], and numerous studies have demonstrated the efficacy of reducing salt for lowering blood pressure [
5] and other risk factors for chronic diseases such as cardiovascular disease (CVD) [
6] and CKD [
7,
8]. Given the high rates of hypertension, CVD and CKD in the Indigenous Australian population, particularly in remote communities [
9], lowering salt intake could significantly reduce chronic disease burden.
Iodine deficiency disorders are a major public health concern, resulting in cognitive impairment, congenital abnormalities, cretinism, hypothyroidism or endemic goitre [
10]. A re-emergence of iodine deficiency was observed in the early 2000’s in Australia, attributable to low usage of iodised table salt and replacement of iodine-rich sanitisers in the dairy industry [
11,
12]. A study conducted in the Northern Territory in 2005–8 found that more than 40 % of Aboriginal teenagers had moderate to severe iodine deficiency (urinary iodine <50 μg/L) [
13]. In 2009, mandatory iodine fortification of bread was introduced in Australia requiring iodised salt be used where salt is added to bread. Current data from the Australian Bureau of Statistics indicate that 11 % of Aboriginal and Torres Strait Islanders are iodine deficient, similar to the non-Indigenous population [
9].
Monitoring population salt and iodine intakes can help to ensure that salt-reduction and iodine deficiency elimination strategies can be implemented synergistically [
10]. The 2012–13 National Aboriginal and Torres Strait Islander Nutrition and Physical Activity Survey (NATSINPAS) was the first national-level dietary survey to report dietary intakes of the Indigenous Australian population [
14]. However discretionary salt intake was not quantified in this survey, meaning sodium and iodine intakes are underestimated. Many remote Indigenous Australian communities have relatively closed food systems, with most of food obtained from the community store, meaning store-sales data can provide an objective indicator of community-level dietary intake over a long time period without burdening individuals in participating communities [
15,
16]. We aim to examine in remote Aboriginal Australian communities using store-sales data, i) estimated adequacy (compared to population-weighted recommendations) and sources of dietary sodium and iodine, and ii) potential effects of strategies to reduce dietary sodium intake on estimated average sodium and iodine intakes. This will inform the development of tailored salt-reduction strategies for this population and identify the potential effects of these strategies on iodine adequacy.
Discussion
We collected 2 years of store-sales data from 20 remote Indigenous community stores representing the main source of food for approximately 8300 individuals as an objective indicator of usual dietary intake. This is the largest apparent consumption dataset for dietary intake specific to those living in remote Indigenous communities to date, and the first study in this population examining the potential effects of salt-reduction strategies on sodium and iodine intakes. We found that estimated average sodium intake was above that recommended while iodine intake was within recommendations. The 2012–2013 NATSINPAS measured self-reported dietary intake of 4100 Indigenous Australians including 2300 living in remote Australia (although did not differentiate between remote and very remote) using 24-h recall. Self-reported dietary sodium and iodine intakes were 2096 mg and 145 μg/day respectively for those living in remote Australia, however these estimates do not include sodium and iodine from discretionary salt, which contributed 19 % of sodium and 21 % of iodine in the present study. Excluding sodium and iodine from discretionary salt, our estimated average intakes (2263 mg Na and 162 μg/I) are within 5 % of the estimates from the NATSINPAS (2378 mg Na and 164 μg I [
14]).
We found that the biggest contributors to salt intake were discretionary salt (20 %), bread (18 %) and processed meat (9 %; ~6 % of which was canned meat). This is consistent with findings from the NATSINPAS; bread and processed meat contributed 20 and 11 % of all sodium for Indigenous Australians in remote setting, (or 16 and 9 % if adjusting for estimated discretionary salt intake) [
14]. Major sources of iodine were bread (35 %), iodised table salt (21 %), milk (19 %) and eggs (6 %). In the NATSINPAS, bread contributed 34 % of iodine, milk 14 % and eggs 5 % (adjusted values 28, 11 and 4 % respectively) for those living remotely.
This analysis highlighted the need for salt-reduction strategies in remote Indigenous communities to reduce salt intakes to within recommended levels. High sodium intakes are prevalent in most populations with access to processed foods [
27], and a similar sodium intake was found when we previously characterised nutrient intake in a smaller number of remote Indigenous communities [
24], therefore the excess sodium intake found in this setting is not surprising. We previously modelled dietary changes to optimise diet quality in remote Indigenous communities, and found that even with large shifts from processed to unprocessed foods, sodium intake remained above the upper limit [
24]. Sodium is ubiquitous throughout the food supply and achieving sustained behaviour change to reduce salt intake is challenging [
27]. As most sodium comes from processed foods, focus has turned to food reformulation as a solution to reduce population salt intake [
27]. In 2009, the Australian Food and Health Dialogue was established and voluntary food reformulation targets were set for maximum sodium levels across a range of commonly consumed foods [
28]. However sodium contents of many products are still above these targets [
26].
Bread is consistently found to be one of the biggest contributors to sodium intake in Australia and other westernised countries, attributable to the moderate-high sodium content of bread and regular consumption [
27]. Sodium content of bread in the present study ranged from 330 to 790 mg Na/100 g for individual products, whereas the target for the maximum salt content of bread set by the Australian Food and Health Dialogue is 400 mg/100 g [
28], suggesting opportunity for further salt-reduction. Modelling reduction in salt content of bread resulted in small reductions in estimated salt intake, similar to those when reduced discretionary salt intake alone was modelled, however either strategy individually could not reduce sodium intake to below the UL. Combining these two strategies could reduce average sodium intake to below the UL, however large (67 %) reductions were needed.
Wide-spread reduction across all processed foods could also reduce estimated salt intakes to within recommendations. Approximately 40 % reduction was needed before the estimated average sodium intake was below the UL. If reduction occurred concurrently with reduced discretionary salt intake, 30 % reduction in both sodium content of processed foods and discretionary salt intake would reduce estimated average sodium intake to below the UL. Such reductions are not unprecedented [
27,
29]. The United Kingdom has achieved reductions in key food products of between 25 and 45 %. This along with a social marketing campaign and labelling changes to target behaviour change to reduce salt intake achieved a 15 % reduction from 9.5 to 8.1 g/day over 10 years [
30]. This was mirrored by blood pressure reduction of 3/1 mm Hg after adjusting for known confounding factors (including fruit and vegetable intake, body mass index and alcohol intake) and a 36 % reduction of both stroke and ischaemic heart disease mortality [
30].
Implementing practical and effective strategies to reduce salt intake is imperative; CVD is the leading cause of mortality in Indigenous Australians [
2], and high blood pressure is one of the strongest predictors of cardiovascular events [
31]. He and Macgregor estimated that a 6 g/day reduction in salt intake (equivalent to 2300 mg sodium) in the UK could reduce systolic/diastolic blood pressure by 5/3 mm Hg reducing risk of stroke by approximately 25 % and ischemic heart disease (IHD) by 18 %, while a 3 g/day (1150 mg sodium) reduction could reduce blood pressure by 2/1 reducing stroke by 13 % and IHD by 10 % [
32]. A reduction in dietary salt as small as 0.5 g/day (200 mg sodium) could have significant outcomes in terms of reduced CVD mortality at the population level [
32]. Since sodium balance is a main role of the kidney, those with kidney impairment may be particularly ‘salt-sensitive’ and therefore more susceptible to the adverse effects of excess salt intake [
7]. Nearly 40 % of the remote Indigenous Australian population have indicators of CKD [
9], therefore population salt-reduction may be particularly beneficial in reducing risk of CVD and CKD progression.
Iodine fortification was an important contributor to iodine intakes in this population as evidenced by half of all dietary iodine coming from the main two targets for iodine fortification (bread and iodised salt). Despite this, iodine was not reduced to below the EAR in any of the modelled scenarios of salt-reduction. This is consistent with previous research showing that salt reduction does not increase iodine deficiency [
10], or that iodine status does not differ across levels of salt intake [
33]. It is important that the messages for salt-reduction and iodine deficiency prevention are consistent [
34]. The WHO recommends this be achieved by: 1) clear non-conflicting messages around iodised salt usage (an example used in Italian public health campaigns is ‘Poco sale, ma iodate!’ which translates to ‘Little salt, but all iodised’), 2) universal salt iodisation (all salt used in processed foods is iodised), and 3) increasing the iodine to sodium ratio in iodised salt as salt intakes are reduced [
10]. Further, the Australian National Health and Medical Research Council (NHMRC) recommended that pregnant and lactating women take daily iodine supplementation [
35].
There are several limitations to the present analysis. Although iodine intake was within recommendations at the population level, we did not measure individual level intake or the proportion of individuals at risk of deficiency; therefore inferences about iodine-sufficiency are applicable at the population-level only. The data composition tables used for this analysis provide an average of like products whereas sodium content may differ by product [
23]. Store-sales data does not account for food wastage (once it has been purchased) or foods obtained from sources other than the community store (or its associated takeaway outlet) such as wild-harvested foods or municipal water. However, it is estimated that this represents only a small proportion of dietary intake [
14]. Wild-harvested foods provided less than 2 % of energy and less than 1 % of sodium and iodine in the NATSNPAS remote Indigenous Australian sample, while water (municipal and bottled combined) provided 4.5 % of iodine and 1.3 % of sodium [
14]. Communities included in the present study did not have food vendors in the community outside of the primary store (except for one community which had a second, smaller store open during the study). The median distance to the nearest competitor was 81.5 km (range 25–443 km).
A further limitation is that the estimated average intakes in this paper were based on an assumption that the average person was consuming sufficient energy to meet the EER, which may not be the case in a population where food insecurity is common [
14]. Average reported energy intake from the NATSINPAS for Indigenous Australians living remotely was 8.5 MJ/person/day, which is 96 % of the population-weighted EER calculated for this study [
14]. Further, the studied population would need to be consuming considerably under the EER for average sodium intake to be within recommendations, and 46 % (36–51 %) of the EER for the average iodine intake to be below the population-weighted EAR. A further limitation of this analysis is that ALPA and OBS have Nutrition Policies regarding availability of iodised versus non-iodised salt (whereby iodised table salt must be stocked in stores where table salt is available), which may limit transferability of the results to other communities.
Acknowledgements
We are grateful to Anthony Gunther, Robyn Liddle, Clare Brown and Rachael Jaenke at Menzies School of Health Research; to SHOP@RIC investigators, project staff and research collaborative; and to community store owners, Arnhem Land Progress Aboriginal Corporation and Outback Stores. The SHOP@RIC study is funded by the Australian National Health and Medical Research Council (NHMRC; ID1024285). EM is supported by a NHMRC Program Grant #631947, and the Australian Primary Health Care Research Institute (supported by a grant from the Commonwealth of Australia as represented by the Department of Health). JB is supported by a National Heart Foundation (NHF) Future Leader Fellowship (ID:100085). JW is supported by a NHMRC/NHF Career Development Fellowship #172121 and receives additional funds from the NHMRC, the World Health Organization (WHO) and the Victorian Health Promotion Foundation (VicHealth) for work on salt reduction. KO is supported by the University of South Australia (UniSA), and is lead investigator on NHMRC Program Grant #631947. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The information and opinions contained in this paper are solely the responsibility of the authors and do not necessarily reflect the views or policy of the Australian Primary Health Care Research Institute, Australian Government Department of Health, NHF, WHO, VicHealth, UniSA or NHMRC.
Competing interests
JW is Director of the World Health Organization Collaborating Centre on Population Salt Reduction.
Authors’ contributions
EM designed the analysis, analysed data and drafted the manuscript. JB conceived and was principal investigator of the SHOP@RIC study, and assisted with design of the analysis, interpretation of the data and drafting the manuscript. JW and KO provided input into interpretation of data and drafting the manuscript. All authors read and approved the final manuscript.