Worldwide and time trends in sodium and potassium intakes in children and adolescents: a systematic review and meta-analysis

Abstract

Background: High sodium (Na) and low potassium (K) intakes in childhood have health effects across the life course. The objective was to estimate global, regional and national Na and K intakes in children since 1990.

Methods: A systematic search of cross-sectional and longitudinal studies measuring Na or K intake in children aged 0-18 years of age since 1990 was conducted. Random effects multilevel meta-analyses and meta-regressions were performed to investigate age and time trends, country and regional differences, and to derive a worldwide average intake.

Results: A total of 259 studies with 520 630 children aged 0-18 years of age (mean 9.7 years) conducted between 1990 and 2021 in 79 different countries (mostly high-income countries) were included. The pooled Na and K intakes were 2.5 g/d (95% CI 2.4, 2.6) and 2.0 g/d (95% CI 1.9, 2.1), respectively. An estimated 73% of children had high Na intake (≥2 g/d/2000 kcal) and 89% had low K intake (<3.5 g/d/2000 kcal). Na intake was the lowest in Sub-Saharan Africa and the highest in North Africa and the Middle East. K intake was the lowest in South Asia and the highest in Central-Eastern Europe and Central Asia. Na and K intakes tended to decrease slightly linearly between 1990 and 2021 and increased logarithmically with age.

Conclusion: Globally, children's Na intake was too high, while K intake was too low. Data were lacking in many countries. Interventions are needed to reduce Na and increase K from childhood, and monitoring should be improved.

Keywords: Blood pressure lowering; Nutrition assessment.

Figure 1. Study selection flow chart. *Some records were excluded for more than one reason.

Figure 2. Data availability and characteristics of included studies. (A) Geographical coverage and representativeness: worldwide map representing the availability of at least national data, subnational data or local data for each country. (B) Countries and regions: tree map representing the number of studies in each country and global burden of disease region (in green: High-income countries, in blue: Latin America and the Caribbean, in pink: Southeast Asia, East Asia and Oceania, in purple: North Africa and the Middle East, in red: Central Europe, Eastern Europe and Central Asia, in brown: Sub-Saharan Africa and in orange: South Asia). (C) Nutrients measured: pie chart representing the number of studies, which measured sodium and potassium intakes (yellow), only sodium (orange) or only potassium (blue). (D) Measurement methods: tree map representing the number of studies using different methods to measure intakes. (E) Age coverage. Histogram representing the number of studies with each mean age. (F) Time coverage: histogram representing the number of studies covering the different calendar years.

Figure 3. Sodium (A) and potassium (B) intakes reported in the included studies. Green-shaded areas indicate recommended intake levels, while red-shaded areas indicate inadequate intakes. Black dotted lines represent meta-analysed mean intakes.

Figure 4. Worldwide estimates for sodium (orange) (g/d/2000 kcal) and potassium (blue) (g/d/2000 kcal). Non-linear metaregression with random effects multilevel model, controlled for age (set at 9 years old) and data collection year (set at 2020).

Figure 5. Country and regional estimates for sodium (orange) and potassium (blue) (g/d/2000 kcal). Non-linear metaregression with random effects multilevel model, controlled for age (set at 9 years old) and data collection year (set at 2020). The dotted lines represent the recommended thresholds (orange: <2 g for sodium, blue: ≥3.5 g for potassium).

Figure 6. Time trends for intakes of sodium (A) and potassium (B). Non-linear metaregression with random effects multilevel model, with covariates age modelled with logarithmic regression and data collection year with linear regression, controlled for age (set at 9 years old).

Figure 7. Age trends for intakes of sodium (A) and potassium (B). Non-linear metaregression with random effects multi-level model, with covariates age modelled with logarithmic regression and data collection year with linear regression, controlled for data collection year (set at 2020).