Effect of South Africa's interim mandatory salt reduction programme on urinary sodium excretion and blood pressure

Abstract

South Africa implemented legislation in June 2016 mandating maximum sodium (Na) levels in processed foods. A pre-post impact evaluation assessed whether the interim legislative approach reduced salt intake and blood pressure. Baseline Na intake was assessed in a nested cohort of the WHO Study on global AGEing and adult health (WHO-SAGE) Wave 2 (Aug-Dec 2015). 24-hour urine samples were collected in a random subsample (n = 1,299; of which n = 750 were considered valid (volume ≥ 300 mL and creatinine ≥ 4 mmol/day (women) or ≥ 6 mmol/day (men))). Follow-up urine samples were collected in Wave 3 (Jun 2018-Jun 2019), with replacements included for those lost to follow-up (n = 1,189; n = 548 valid). In those aged 18 - 49y, median salt intake was 7.8 (4.7, 12.0) g/day in W2 (n = 274), remaining similar in the W3 sample (7.7 (4.9, 11.3) g salt/day (n = 92); P = 0.569). In older adults (50 + y), median salt intake was 5.8 (4.0, 8.5) g/day (n = 467) in W2, and 6.0 (4.0, 8.6) g/day (n = 455) in W3 (P = 0.721). Controlling for differences in background characteristics, overall salt intake dropped by 1.15 g/day (P = 0.028). 24hr urinary Na concentrations from a countrywide South African sample suggest that salt intakes have dropped during the interim phase of mandatory sodium legislation. Further measurement of population level salt intake following stricter Na targets, enforced from June 2019, is necessary.

Keywords: Blood pressure; Food policy; Legislation; Potassium; Salt reduction; South Africa.

Fig. 1

Flowchart of recruitment and data available from Waves 2 and 3: Urine samples, blood pressure and questionnaire (CAPI) data.

Fig. 2

Salt intake between Wave 2 and Wave 3, according to age, sex, obesity and hypertension status. Box plots represent median and IQR. Independent samples with valid 24hr urine collections: Wave 2 (n = 750, blue) and Wave 3 (n = 548, red). P-values are from Mann–Whitney tests for differences between the waves. Analyses were adjusted for: age, ethnicity, location, marital status, education level, alcohol use, smoking status, waist-hip ratio, hypertension status at W3, SBP and DBP except if the variable was the categorizing factor. BMI: Body Mass Index; HTN: hypertensive; no HTN: not hypertensive. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)