The impact of the reduction of sodium content in processed foods in salt intake in Brazil

This study aimed at assessing the potential impact of the reduction of sodium content in processed foods in the average salt intake in the Brazilian population. A total of 32,900 participants of the first National Dietary Survey (NDS 2008-2009), age 10 years and older who provided information about food intake over two days were evaluated. The sodium reduction targets established by the Brazilian Ministry of Health in 2010 and 2013 were used as the reference to determine the maximum content of sodium in 21 groups of processed food. The results show that sodium reduction targets in processed food have small impact in mean Brazilian population intake of salt. For 2017, the expected mean reduction is of 1.5%, the average sodium intake being still above the recommended 2,000mg/ day maximum. Therefore, it will hardly be possible to reach the necessary reduction in salt intake in Brazil from volunteer agreements like the ones made so far. Sodium; Dietary Sodium; Industrialized Foods; Diet Surveys http://dx.doi.org/10.1590/0102-311X00064615 Correspondence A. de Moura Souza Instituto de Estudos em Saúde Coletiva, Universidade Federal do Rio de Janeiro. Av. Horácio Macedo s/n, Rio de Janeiro, RJ 21941-598, Brasil. amandamoura@msn.com de Moura Souza A et al. 2 Cad. Saúde Pública, Rio de Janeiro, 32(2):e00064615, fev, 2016 Introduction Excessive salt intake is associated with a progressive increase of blood pressure, and therefore to an increase in the risk of cardiovascular diseases, which are the main cause of deaths in Brazil and worldwide 1,2. In 2013, the prevalence of self-reported high blood pressure in the Brazilian population was 24% 3. According to data from the first National Dietary Survey (NDS) conducted in 2008-2009, the average salt intake among the Brazilian population was 3,190mg/day4, which is higher than the maximum tolerable level of 2,000mg/day for adults, as recommended by the World Health Organization (WHO) 5. According to Sarno et al. 6, the intake of processed foods accounted for a mean salt intake increase from 17.2% to 20.5%. In Brazil, the Ministry of Health has been coordinating national strategies aimed at reducing sodium intake, in addition to making agreements with the food industry for a gradual reduction in salt levels in the processed food categories considered as priority 7,8,9,10,11. The aim of this investigation was to assess the potential impact of sodium content reduction in processed foods on the overall sodium intake in Brazil.


Introduction
Excessive salt intake is associated with a progressive increase of blood pressure, and therefore to an increase in the risk of cardiovascular diseases, which are the main cause of deaths in Brazil and worldwide 1,2 .In 2013, the prevalence of self-reported high blood pressure in the Brazilian population was 24% 3 .
According to data from the first National Dietary Survey (NDS) conducted in 2008-2009, the average salt intake among the Brazilian population was 3,190mg/day4, which is higher than the maximum tolerable level of 2,000mg/day for adults, as recommended by the World Health Organization (WHO) 5 .According to Sarno et al. 6 , the intake of processed foods accounted for a mean salt intake increase from 17.2% to 20.5%.
In Brazil, the Ministry of Health has been coordinating national strategies aimed at reducing sodium intake, in addition to making agreements with the food industry for a gradual reduction in salt levels in the processed food categories considered as priority 7,8,9,10,11 .The aim of this investigation was to assess the potential impact of sodium content reduction in processed foods on the overall sodium intake in Brazil.

Methods
This study used data from the NDS, conducted in 2008-2009, in a subsample of 24% of the 55,790 families surveyed in the 2008-2009 Household Budget Survey conducted by the Brazilian Institute of Geography and Statistics (IBGE) 12 .
The goal of the NDS was to collect food-intake data of all residents of the selected households, age 10 years and over.Food intake was estimated from the average intake recorded in two nonconsecutive days.A total of 32,900 NDS participants who provided data of their intake over two days were assessed.
The amount of grams of food intake and their sodium content in milligrams (mg) were estimated with the use of the  12 .
Salt reduction targets, agreed upon by the Brazilian Ministry of Health and the food industry associations, in 2011, to be reached the at completion of five years after the agreement, were used as the reference to determine the maximum amount of sodium in processed food categories considered a priority.The criteria were: (1) international references about sodium reduction; (2) lower limit of sodium content in the respective category, in the domestic market; (3) adjusted means of sodium levels and/or amount found in at least 50% of the products of the category found in the domestic market 7,8,9,10,11 (Table 1).
For the targets to be found, initially the mean amount of sodium (mg Na/100g of food) was calculated for each of the 21 food groups or items considered a priority.The sum of sodium content in each food group was divided by the amount of grams intake multiplied by 100.Mean sodium intake estimates (mg/day) were calculated for the years 2013, 2016 and 2017, taking in consideration the agreed sodium content reduction, with the overall average reduction proportion calculated at the completion of five years after the agreement.Afterward, for 2017, a further 25% reduction was added to the food whose maximum sodium content agreed upon was higher than the mean estimated by the 2008-2009 NDS, and to the food whose agreed reduction was lower than 25%.This proportion was chosen because, at this salt reduction level, no change in flavor is detected by the consumer 13 .The analyses were stratified according to sex and age group (10-19 years, 20-59 years, and 60 years and older).
All analyses took into consideration the sample weight and the study design, and were performed using SAS version 9.1.3software (SAS Inst., Cary, USA).
The NDS was approved by the Ethics Research Committee of the Social Medicine Institute, State University of Rio de Janeiro, registry filing number CAAE 0011.0259.000-11.

Results
The maximum content of sodium agreed upon was higher than the estimated amount based on the food composition table, except for filling cookies, savory biscuit, bread loaf, breakfast cereal, mayonnaise, ham, French fries, sweet biscuit, soups, and French roll (Table 1).
The estimated mean sodium intake reduction 5 years after the agreement was 1.5%.After applying the 25% reduction for the year 2017, the estimated mean reduction was 6.3% (Table 2).The mean sodium intake reduction was similar among the sex and age groups, ranging between 5.6% among male adults and 7.6% among women.

Discussion
The goals of salt reduction in processed foods have small impact in the average sodium intake by the Brazilian population, which remains above Cad.Saúde Pública, Rio de Janeiro, 32(2):e00064615, fev, 2016 * Aplying the 25% sodium content reduction in the food groups whose maximum sodium content agreed upon was higher than the estimated in the National Dietary Survey (NDS) and in those whose proportion of reduction was lower than 25%.
the maximum recommended limit of 2,000mg/ day.The estimated reduction is below the 20% observed by Grimes et al. 14 , who assessed the reduction of sodium in processed foods in the average salt intake of Australian children and adolescents.However, the food intake standards of countries like Australia are quite different than Brazil's, and that should be taken into account when assessing the impact of these policies.In addition, a possible explanation for our findings is the fact that higher proportion of reduction had been agreed upon for food with lower sodium content.For instance, for the filling cookies, with sodium density of 405mg/100g, the total reduction proportion agreed upon was 34.6%.However, when compared to ham, whose sodium density is 1,313mg/100g, about three times higher than the cookie's, the agreed reduction was 11.6%, around three times lower.Most of the reduction amounts agreed upon were higher than the mean sodium content estimates presented in the 2008-2009 NDS nutritional composition table.This fact may be explained by differences between the brands of food included in the table and the brands analyzed to determine the mean content agreed upon.Moreover, there is no knowledge about the market share of the brands of food included in the agreements.
In 2013, WHO established a 30% global reduction goal for the total sodium intake by 2025, and one of the key actions is to reduce the content of sodium in processed products 15 .Some 17 countries have developed volunteer or legally imposed sodium content reduction in processed foods, with bread being the main target of these actions 16 .In Brazil, the agreed proportion of salt content reduction for the French roll was low, despite its high sodium density and high prevalence of consumption (it is the third type of food eaten the most in Brazil) 4,17 .For the French roll, the total reduction was about 10%.However, a 55% reduction in the sodium content of this type of food, considering the lowest figure found in an American food composition table 18 , would lead to a 9% reduction in the mean salt intake (data not shown).Of note is the fact that food with low sodium density, but eaten more frequently, like rice and beans, are among those that contribute the most for the intake of sodium among the Brazilian population 4 .Despite the contribution of processed foods to the mean sodium intake, the contribution of table salt and salt-based seasoning is noteworthy 6 .Therefore, in Brazil, it will be difficult to achieve the necessary sodium intake reduction with voluntary agreements like the ones that were made to date.Considering the increase of intake of ultraprocessed food, reviewing the targets or including other groups of foods will likely not suffice.For that, public policies that encourage behavioral changes towards reducing the use of salt in the preparation of food, and lower intake of ultraprocessed foods, based on the new Dietary Guidelines for the Brazilian Population 19 play a crucial role in effectively changing sodium intake standards in the Brazilian population.

Contributors
A. de Moura Souza contributed to the data analysis and interpretation, writing of the article, critical review of the academic content and approval of the final version to be published.B. S. N. Souza and I. N. Bezerra contributed to the data analysis and interpretation, critical review of the academic content and approval of the final version to be published.R. Sichieri contributed to the data design, project and interpretation, critical review of the academic content and approval of the final version to be published.The journal has been informed about some errors in the paper.The corrections are follows: A revista foi informada sobre alguns erros no artigo.As correções seguem abaixo: La revista fue informada sobre algunos errores en el artículo.Siguen las correcciones: • On page 2, second column, 6 th paragraph, line 2, where the text reads: After applying the 25% reduction for the year 2017, the estimated mean reduction was 6.3% (Table 2).The mean sodium intake reduction was similar among the sex and age groups, ranging between 5.6% among male adults and 7.6% among women.it shoud read: After applying the 25% reduction for the year 2017, the estimated mean reduction was 9.2% (Table 2).The mean sodium intake reduction was similar among the sex and age groups, ranging between 8.5% among male adults and 10.4% among aged women.
• On Table 1, where the text reads: it shoud read: • On Table 2, where the text reads: Table 2 Impact estimates of the processed food sodium content reduction in the mean intake of sodium (mg/day), standard error (SE) and mean reduction of the proportion of sodium intake in 2017, according to sex and age group.* Aplying the 25% sodium content reduction in the food groups whose maximum sodium content agreed upon was higher than the estimated in the National Dietary Survey (NDS) and in those whose proportion of reduction was lower than 25%.* Aplying the 25% sodium content reduction in the food groups whose maximum sodium content agreed upon was higher than the estimated in the National Dietary Survey (NDS) and in those whose proportion of reduction was lower than 25%.

2008-
Table of Reference Measures and Table of Nutritional Composition of Food Consumed in Brazil, respectively, both developed by the IBGE

Table 1
Mean sodium content, maximum content of sodium agreed for the years 2013, 2016 and 2017, and final proportion of reduction at the end of five years after the agreement, for the food groups assessed by the National Dietary Survey(NDS 2008-2009).

Table 2 Impact
estimates of the processed food sodium content reduction in the mean intake of sodium (mg/day), standard error (SE) and mean reduction of the proportion of sodium intake in 2017, according to sex and age group.

Souza BSN, Bezerra IN, Sichieri R. The impact of the reduction of sodium content in processed foods in salt intake in Brazil. Cad Saúde Pública 2016; 32(2):e00064615
Cad. Saúde Pública, Rio de Janeiro, 32(2):e00064615, fev, 2016Cad.Saúde Pública 2016; 33(1):eER011116This article is published in Open Access under the Creative Commons Attribution license, which allows use, distribution, and reproduction in any medium, without restrictions, as long as the original work is correctly cited.

Table 1
Mean sodium content, maximum content of sodium agreed for the years 2013, 2016 and 2017, and final proportion of reduction at the end of five years after the agreement, for the food groups assessed by the National Dietary Survey

Table 1
Mean sodium content, maximum content of sodium agreed for the years 2013, 2016 and 2017, and final proportion of reduction at the end of five years after the agreement, for the food groups assessed by the National Dietary Survey(NDS   2008-2009).

Table 2
Impact estimates of the processed food sodium content reduction in the mean intake of sodium (mg/day), standard error (SE) and mean reduction of the proportion of sodium intake in 2017, according to sex and age groups.