Table 1.

IMPACT_NCD-BR data sources

Parameter	Outcome	Details	Comments	Source
Population size estimates	Population	Brazilian Institute of Geography and Statistics (IBGE)	Stratified by age and sex	Brazilian Institute of Geography and Statistics (IBGE) [17]
Population projections	Population	2010–2060 Brazil population projections produced by IBGE	Stratified by year, age and sex	Brazilian Institute of Geography and Statistics (IBGE) [18]
Mortality	Deaths from CHD, stroke, and any other non-modelled causes	Underlying cause of death 2000–2016	Stratified by year, age and sex	National Mortality Information System (SIM/SUS) [19]
Sodium exposure	Exposure of individuals	National Household Budgetary Survey	Anonymised, individual-level dataset	IBGE - National Household Budgetary Survey (POF) 2008–2009 [20, 21]
Systolic blood pressure exposure	Exposure of individuals	National Health Survey	Anonymised, individual-level dataset	IBGE - National Health Survey (PNS) 2013 [22]
Effect of sodium consumption on systolic blood pressure	Systolic blood pressure	Meta-analysis/meta-regression of 103 trials	Only trials with duration >7 days were analysed	Mozaffarian et al. [3]
Reference level of sodium consumption	Ideal sodium consumption below which no excess risk was considered to occur	Evidence from ecological studies, randomised trials, and meta-analyses of prospective cohort studies	Intake levels associated with the lowest risk ranged from 614 to 2391 mg/day; in large, well-controlled randomised feeding trials, the lowest tested intake for which blood pressure reductions were clearly documented was 1500 mg/day	Mozaffarian et al. [3]
Relative risk for systolic blood pressure	CHD and stroke incidence (ICD-10: I20–I25 and I60–I69)	Pooled analysis of 2 individual-level meta-analyses	Stratified by age and sex; adjusted for regression dilution and total blood cholesterol and, where available, lipid fractions (HDL and non-HDL cholesterol), diabetes, weight, alcohol consumption, and smoking at baseline	Micha et al. [23]
	Mortality from any cause excluding CHD and stroke	Individual-level meta-analysis of 48 prospective cohort studies	Adjusted for age, sex, race or ethnicity, deprivation, smoking, diabetes, inactivity, alcohol, and obesity	Stringhini et al. [24]
Reference level of systolic blood pressure	Ideal systolic blood pressure below which no excess risk was considered to occur	Evidence from randomised trials of antihypertensive drugs and the INTERSALT study	There may be health benefits by lowering systolic blood pressure down to 110 mm Hg	Singh et al. [25]
Disease costs	Public hospitalisation costs for CHD and stroke	Underlying cause of hospitalisation (2018)	Average cost of hospitalisations per individual	National Hospital Information System (SIH/SUS) [26]
	Primary health, outpatient and informal care and medication costs for CHD and stroke		Costs were extrapolated to Brazilian settings	Leal et al. [27]