Potassium: the mineral most people under-eat, and the one some people must limit

Potassium is the rare nutrient where the average person eats too little and a specific group has to deliberately eat less. What separates the two situations is kidney function.

What it does in the body

• Potassium maintains fluid volume inside cells and the electrical gradients needed for nerve transmission, muscle contraction and kidney function. Evidence E source

How much you need

• The Adequate Intake for potassium is 3,400 mg a day for adult men and 2,600 mg for adult women. — adults 19–50 Evidence E source
• There is no RDA for potassium because the data were insufficient to derive one, and no upper limit because high intakes do not cause hyperkalemia in adults with normal kidney function. — adults with normal kidney function Evidence E source
• Average potassium intake in the United States is 3,016 mg a day for men and 2,320 mg for women, below the recommended amount. — US adults 20+ Evidence C source
• More potassium is not simply better for blood pressure: the curve bends, the benefit fades above roughly 30 mmol a day extra, and past about 80 mmol a day pressure starts going back up. — adults, mostly with hypertension and mostly in crossover trials, given supplemental potassium at doses from 30 to 140 mmol a day Evidence A source
• In prospective studies, each extra gram of potassium a day from food went with about 15 per cent less cardiovascular disease and 7 per cent lower death rates. — adults followed prospectively, with potassium assessed as dietary intake, blood level or urinary excretion Evidence C source
• What lowers blood pressure is the balance between sodium and potassium, not either one alone: trials that shifted the ratio downwards got about 1 mmHg more off systolic pressure. — adults in 5 randomised trials out of 39 studies reviewed, with 24-hour urine collections Evidence A source
• Cutting salt works about as well whether you eat a lot of potassium or a little: across 68 trials, starting potassium intake did not change the result. — adults in sodium-reduction trials; median baseline potassium intake 67.7 mmol a day (interquartile range 54.6 to 76.4) and mean sodium reduction 128 mmol Evidence A source
• A low sodium salt is not one product. Across 87 of them sold worldwide, potassium chloride ran from none at all to the entire contents of the packet. — 87 low sodium salt products found on sale in 47 of the world's 195 countries Evidence C source
• How much potassium a vegetable carries depends on the dirt it grew in: across 140 studies of pepper, fruit grown at soil pH 6.5 to 7.5 came out richer in potassium than fruit from acid soil. — 140 publications on pepper fruit, giving 649 datapoints for potassium Evidence C source
• Fruit and vegetables supply a fifth of the potassium Americans actually eat. Milk alone supplies more than half of what vegetables do. — United States population, What We Eat In America / NHANES 2009-2010 Evidence E source
• The body that tells the world to eat more potassium says in the same breath that it does not mean tablets or salt substitutes. Its figure is at least 3510 mg a day, and it means food. — adults in the general population Evidence E source

How well your body absorbs it

• Potassium from a supplement is absorbed slightly better than potassium from an ordinary diet, and the gap is about three and a half percentage points. — 30 pre-hypertensive to hypertensive adults, 15 men and 15 women, mean age 48 years Evidence B source
• The form of potassium changed what the kidney did with calcium: the gluconate supplement cut urinary calcium and made urine less acid, and the same amount of potassium from potatoes did neither. — 30 pre-hypertensive to hypertensive adults, mean age 48.2 years, mean body mass index 31.4 Evidence B source
• Cooking method decides how much potassium is left on the plate. Pressure cooked spinach came in at 74.85 mg per 100 g, while the same vegetables baked dry stayed far above the level a kidney diet allows. — foods chosen from the Brazilian TACO table for containing above 201 mg of potassium per 100 g. 68 heart disease patients rated each food Evidence B source
• Two groups ate the same 350 g of vegetables a day for two weeks. Only the group whose vegetables were cooked without water, in multi-layer pans, raised the potassium coming out in their urine. — adults in a two week intervention with health examinations at both ends Evidence B source

What too little does

• Mild low potassium causes constipation, fatigue, muscle weakness and malaise. Severe cases cause paralysis, poor respiration and cardiac arrhythmias. — people with hypokalemia Evidence E source
• On days when potassium dipped below 3.5 mmol/L, atrial fibrillation was more than twice as likely to appear in people wearing an implanted monitor. — participants with stroke risk factors carrying an implantable loop recorder in the LOOP study Evidence C source
• A blood test tells you the opposite story from the dinner plate: in the same review, higher potassium in the blood went with 23 per cent higher mortality while higher potassium in the urine went with 16 per cent lower. — adults followed prospectively, comparing three different ways of measuring potassium status Evidence C source

Who is most likely to fall short

• Among people who already have heart disease, a high potassium reading in hospital went with nearly three times the risk of dying there, though the excess shrinks the further out you look. — patients with cardiovascular disease such as myocardial infarction, stroke and heart failure, average age 68.3 years Evidence C source
• The standard advice to cut potassium in kidney disease rests on very low-quality evidence: restricting it moved blood potassium by 0.22 millimoles per litre and did not slow the disease. — adults with chronic kidney disease Evidence C source

What the evidence does and does not show

• A 2017 meta-analysis of 25 randomised trials in people with hypertension found potassium supplementation lowered systolic blood pressure by 4.48 mmHg and diastolic by 2.96 mmHg. — 1,163 participants with hypertension Evidence A source
• In 247,510 adults across 11 prospective cohorts, 1,640 mg a day more potassium was associated with 21% lower stroke risk. — 247,510 adults, prospective cohorts Evidence C source
• In people whose blood pressure is already normal, potassium supplements lower systolic pressure by about 2 mm Hg and leave diastolic pressure unchanged. — adults of the general, normotensive population, potassium arm of a six-supplement review Evidence A source
• Replacing ordinary salt with a potassium-based salt substitute lowers systolic pressure by about 5 mm Hg and diastolic by about 2 mm Hg in adults. — adults in 26 trials, 11 of them in people with hypertension, 11 mixed, one in normotensive and one in pre-hypertensive participants; 14 of 26 trials in low- and middle-income countries Evidence A source
• The same Cochrane review found the salt substitute also raised blood potassium slightly, and that clinical hyperkalaemia did not become more common. — adults in the same 26 trials, all of which excluded people at known risk from extra potassium Evidence A source
• In the largest trial ever run on this, swapping a quarter of household salt for potassium chloride cut strokes by 14 per cent and deaths from any cause by 12 per cent over almost five years. — rural Chinese adults with a history of stroke or aged 60 and over with high blood pressure; mean age 65.4 years, 72.6% with prior stroke, 88.4% with hypertension Evidence B source
• Among older adults who started with normal blood pressure, a potassium-enriched salt substitute cut the rate of newly developing hypertension by about 40 per cent over two years without causing more low-pressure episodes. — older Chinese adults with normal blood pressure at baseline living in elderly care facilities, followed for two years Evidence B source
• Potassium supplements lower systolic pressure by about 3.7 mm Hg across mixed adult populations, while also raising plasma aldosterone, a hormone that pushes pressure the other way. — adults in 36 trials meeting the inclusion criteria, of which 18 also reported systolic blood pressure Evidence A source
• In trials that deliberately raised potassium intake by swapping ordinary salt, serious adverse events were no more frequent than on regular salt, but the certainty of that reassurance is very low. — adults, mostly aged around 62 and at higher-than-average cardiovascular risk, in trials mainly conducted in China or Taiwan Evidence A source
• The body held on to almost twice as much potassium from a supplement as from the same amount eaten as potatoes, and it was the potatoes that moved blood pressure. — 30 pre-hypertensive to hypertensive adults on fully controlled 16-day diets with two week washouts Evidence B source
• WHO published its salt substitute advice together with a list of what it still does not know, and the first gap on that list is ordinary people: the trials were run on the hypertensive and the high risk. — the general population, children and pregnant women, none of whom the evidence covers Evidence E source
• The same extra potassium moves blood pressure ten times further in someone who already has hypertension: about 5.3 mmHg of systolic against 0.5 mmHg in someone whose pressure is normal. — 684 adults in 10 randomised trials published between 2005 and 2021, in the UK, USA, Italy, China, the Netherlands and Denmark. Six groups had hypertension, four did not Evidence A source
• Everything we can say about how much potassium moves blood pressure rests on 684 people studied for between one and six weeks. — the 10 randomised trials that survived selection out of 1826 identified references Evidence A source

What too much does

• In people whose kidneys cannot clear potassium, even intakes below the recommended amount can cause dangerous hyperkalemia. — people with chronic kidney disease or on certain drugs Evidence E source
• The safest blood potassium is between 4 and 4.5 millimoles per litre. At 5.5 the risk of dying is 22 per cent higher, and being too low is worse still. — general population, high cardiovascular risk and chronic kidney disease cohorts; average age 55, average eGFR 83 mL/min/1.73 m2, mean baseline potassium 4.2 mmol/L Evidence C source
• A common antibiotic pushes one person in five above 5 millimoles per litre within a week, which shows the dangerous dose of potassium usually arrives from a prescription rather than a plate. — patients treated with trimethoprim, in 18 retained reports Evidence C source
• Across every trial WHO could find, swapping in a potassium salt substitute did not meaningfully raise the rate of high blood potassium, over follow-up reaching five years. — randomised trials of lower sodium salt substitutes in adults Evidence E source
• A potassium tablet is capped at 99 mg by custom, not by law, and 99 mg is about 2 percent of the Daily Value. No one can top up a diet with these. — dietary supplements sold in the United States Evidence E source
• There is no upper limit for potassium from food, and the reason is not safety but absence: the committee looked at the case reports and decided they were not enough to set one. — adults with normal kidney function Evidence E source

Medicines it interacts with

• ACE inhibitors, ARBs and potassium-sparing diuretics reduce urinary potassium excretion and can cause hyperkalemia. — people on these medications Evidence E source
• Pairing a thiazide with a potassium-sparing diuretic lowers blood pressure more than the thiazide alone, and it is the thiazide alone that drains the most potassium. — patients with primary hypertension, mean age 55 years, 45 per cent women, across 276 trials Evidence A source
• The dose is what decides it: 25 mg of hydrochlorothiazide added about 2.5 low-potassium events per hundred people over eight weeks, while 12.5 mg added none that could be measured. — patients with uncomplicated hypertension on angiotensin receptor blockers, a calcium channel blocker and hydrochlorothiazide as single drugs or fixed-dose combinations, median follow-up 8 weeks Evidence A source
• Taking potassium chloride alongside a thiazide did not undo the diuretic's effect on blood sugar, and it pushed uric acid higher than the diuretic alone. — hypertensive patients randomised equally to the two arms Evidence B source
• Adding a thiazide on top of intravenous furosemide in acute heart failure multiplied the odds of low potassium roughly fivefold. — patients hospitalised with acute heart failure and volume overload, randomised to hydrochlorothiazide or placebo on top of intravenous furosemide Evidence B source
• Potassium runs the other way with some drugs: SGLT2 inhibitors cut the risk of dangerously high potassium by about a sixth without pushing anyone low. — people with type 2 diabetes at high cardiovascular risk or with chronic kidney disease, across 6 trials of 4 different SGLT2 inhibitors Evidence A source
• The reassurance about salt substitutes was built on people who were screened to be safe. Every trial in the WHO review threw out anyone whose kidneys or medicines put them at risk. — participants of the trials included in the WHO systematic review Evidence E source
• WHO attaches a condition to salt substitutes that has nothing to do with the salt: use them where kidney disease gets diagnosed, and monitor. — national policy makers and programme managers Evidence E source

What to do with this

Read the caution before the food list, because this page has two audiences and one of them should do the opposite of the other.

If your kidneys work normally, eat more of it and stop counting. Potatoes with the skin, beans, lentils, tomatoes, cooked greens, avocado, dairy and fish. It accumulates across a day of whole food and it never accumulates across a day of packaged food, which is most of the story of why average intake falls short.

If you have chronic kidney disease, or take an ACE inhibitor, an ARB or a potassium-sparing diuretic, none of that applies to you. Watch out for salt substitutes in particular, because most of them are potassium chloride and they are marketed as the healthy option. Your number comes from your clinician.

If a blood test brought you here, the comfortable range is between 4 and 4.5 millimoles per litre, and a low reading is worse than a high one. A high one is more often a question about kidneys or about a prescription than about your diet. One commonly used antibiotic pushes one person in five above 5 millimoles per litre within a week.

If you have chronic kidney disease and have been told to cut potassium, that is worth asking about rather than following in silence. Pooling the trials, restricting potassium moved blood potassium by 0.22 millimoles per litre and did not slow the disease, and the evidence behind it was rated very low quality. Your clinician still sets your number. Ask how tight it has to be, because a diet with no fruit and no vegetables in it costs you something too.

Two bodies currently publish two numbers, and neither is wrong. The United States adequate intake is 3,400 mg for men and 2,600 for women. WHO suggests at least 3,510 mg for adults and says in the same breath that it does not mean supplements in tablet form or salt substitutes. If you have been told you fall short, check which number you were measured against before you do anything about it.

If a salt substitute is the plan, read the packet rather than the category. Across 87 low sodium salt products on sale in 47 countries, potassium chloride ranged from none of the packet to all of it, and only half carried any label about health risks. Two products with the same shelf position can hand you completely different amounts of potassium.

Cooking changes what reaches the plate, and the direction is the one people forget. Two groups ate the same 350 g of vegetables a day for two weeks. Only the group whose vegetables were cooked without water raised the potassium coming out in their urine, from 1.6 to 2.0 grams. Their sodium to potassium ratio fell from 3.7 to 2.6 and nobody else's moved. The trial cannot separate a pot that leaches less from a pot that makes the food nicer to eat, and it calls itself double blind while serving visibly different food, so hold the size of it loosely.

Where to get it from food

The bottom line

For most people, eating more potassium-rich food is one of the better-supported changes you can make, because the blood pressure effect has been measured in randomised trials and the stroke association is large. For people with chronic kidney disease, or on ACE inhibitors, ARBs or potassium-sparing diuretics, that same advice is dangerous. This is why general nutrition advice should never be applied without knowing which group you are in. Two findings have landed since this page was written. The first is that more is not automatically better. Pooling the trials, the blood pressure benefit fades above roughly 30 mmol a day of extra potassium, and past about 80 mmol a day pressure starts climbing again, so the whole gain sits in the first step out of a low intake. The second is the salt substitute, which works on sodium and potassium at the same time. Replacing a quarter of household salt with potassium chloride cut strokes by 14 per cent and deaths from any cause by 12 per cent over almost five years. In older adults who started with normal pressure, the same swap cut the rate of newly developing hypertension by about 40 per cent. That is the largest effect anything on this page has produced. It is also the one product that can put somebody whose kidneys cannot clear potassium in hospital. Serious harms in those trials came out no more frequent than on ordinary salt, at very low certainty, which is the absence of a known problem rather than proof of safety. Three things published since sharpen the picture rather than change it. The first is who the blood pressure benefit belongs to. Pooling ten randomised trials in 684 adults, where potassium was measured in 24-hour urine rather than reported, a rise of 50 mmol a day lowered systolic pressure by 5.3 mmHg in people who already had hypertension and by 0.5 mmHg in people who did not. That is close to the whole effect sitting in one group. Those trials ran one to six weeks and the two groups were given different amounts of extra potassium, so it is not a like for like comparison. The second is that a potassium tablet cannot deliver any of this. Manufacturers hold supplements at 99 mg a unit, about 2 per cent of the Daily Value, over concerns tied to potassium-containing drugs. The third is that no upper limit for potassium exists at all. The committee looked and decided the evidence would not support setting one, which is a statement about missing evidence rather than a licence, and the same fact sheet notes that very large supplement doses have caused heart abnormalities and death in case reports.

Sources

1. k-01 · Position of an expert body · expert body statement
   Potassium is required for normal cell function because of its role in maintaining intracellular fluid volume and transmembrane electrochemical gradients.
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · extracted
2. k-02 · Position of an expert body · expert body recommendation
   For adults 19–50 years: 3,400 mg for males and 2,600 mg for females. (value from the source's table, not a sentence)
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · independently re-checked
3. k-03 · Position of an expert body · expert body statement
   There is no evidence that high intakes of potassium cause hyperkalemia in adults with normal kidney function or other adverse effects. Therefore, the committee did not set a UL.
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · independently re-checked
4. k-04 · Observational data · national survey data
   In adults age 20 and over, the average daily potassium intake from foods is 3,016 mg for men and 2,320 mg for women.
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · independently re-checked
5. k-r2-1 · Meta-analysis or systematic review · dose-response meta-analysis of randomised trials
   We observed a U-shaped relationship between 24-hour active and control arm differences in potassium excretion and BP levels, with weakening of the BP reduction effect above differences of 30 mmol/d and a BP increase above differences ≈80 mmol/d.
   Filippini et al., Journal of the American Heart Association, 2020 · checked 2026-09-17 · independently re-checked
6. k-r2-2 · Observational data · meta-analysis of prospective observational studies
   Each 1.0 g/d increment in potassium intake was associated with a decreased risk of CVD (RR: 0.85, 95% CI: 0.80-0.91) and all-cause mortality (RR: 0.93, 95% CI: 0.88-0.99).
   Fan et al., Critical Reviews in Food Science and Nutrition, 2024 · checked 2026-09-17 · independently re-checked
7. k-r2-4 · Meta-analysis or systematic review · systematic review and meta-analysis of randomised trials
   Meta-analysis of 5 RCTs found a lower UNa: K ratio to be associated with a significantly greater reduction in systolic and diastolic BP compared with a higher UNa: K ratio [SMD: -1.09 (95% CI: -1.91, -0.28) mmHg and -1.42 (95% CI: -2.24, -0.59) mmHg, respectively].
   Ndanuko et al., Advances in Nutrition, 2021 · checked 2026-09-17 · independently re-checked
8. k-r2-5 · Meta-analysis or systematic review · systematic review and meta-analysis of randomised trials
   The median intake of baseline potassium was 67.7 mmol (Interquartile range: 54.6-76.4 mmol), and the mean reduction in sodium intake was 128 mmol (95% CI: 107-148)...did not identify a significant association of baseline potassium intake levels with the blood pressure reduction achieved with a 50 mmol lowering of sodium intake (p > 0.05 for both systolic and diastolic blood pressure).
   Huang et al., Journal of Human Hypertension, 2021 · checked 2026-09-17 · independently re-checked
9. k-e3-01 · Observational data · systematic environmental scan of products
   The proportion of sodium chloride varied from 0% (sodium-free) to 88% (as percent of weight; regular salt is 100% sodium chloride). Potassium chloride was the most frequent component with levels ranging from 0% to 100% (potassium chloride salt).
   Yin et al., JMIR Public Health and Surveillance, 2021 · checked 2026-09-18 · independently re-checked
10. k-e3-02 · Observational data · meta-analysis of composition measurements
    Pepper fruits produced at pH 6.5-7.5 had higher fruit K concentration while acidic soils (pH<6.5) favored higher Fe and Zn concentrations.
    Food Research International, 2024 · checked 2026-09-18 · independently re-checked
11. k-e3-08 · Position of an expert body · cross-sectional dietary survey brief
    As shown in Table 3, Fruits and Vegetables contributed 20% total potassium intake, providing 13% and 7% respectively.
    FSRG Dietary Data Briefs, USDA Agricultural Research Service, WWEIA/NHANES 2009-2010 · checked 2026-09-18 · independently re-checked
12. k-e3-09 · Position of an expert body · guideline recommendation
    WHO suggests a potassium intake of at least 90 mmol/day (3510 mg/day) for adults ... This recommendation does not include obtaining potassium from supplements in tablet form or LSSS to increase potassium intake.
    Use of lower-sodium salt substitutes: WHO guideline, World Health Organization, 2025 (CC BY-NC-SA 3.0 IGO) · checked 2026-09-18 · independently re-checked
13. k-e3-03 · Randomised controlled trial(s) · randomised crossover controlled feeding trial
    For potassium, the %ab in the supplement phase (91.9 ± 1.0) was significantly higher compared to the control (88.3 ± 1.3; p = 0.003).
    Stone et al., Nutrients, 2021 · checked 2026-09-18 · independently re-checked
14. k-e3-05 · Randomised controlled trial(s) · randomised crossover controlled feeding trial, secondary analysis
    Using a mixed model ANOVA there was a significantly lower urinary Ca excretion in the K-gluconate phase (96 ± 10 mg/day) compared to the control (115 ± 10 mg/day; p = 0.027) and potato (114 ± 10 mg/day; p = 0.033).
    Stone et al., Nutrients, 2021, вторинний аналіз · checked 2026-09-18 · independently re-checked
15. k-e3-06 · Randomised controlled trial(s) · randomised crossover clinical trial with food assay
    The vegetables with the lowest potassium values were spinach (74.85 mg/100 g) and zucchini (93.62 mg/100 g), both cooked under pressure.
    Clinical Nutrition ESPEN, 2025 · checked 2026-09-18 · independently re-checked
16. k-e3-07 · Randomised controlled trial(s) · randomised controlled trial
    The amount of K excreted daily into urine (calculated as K) increased significantly after the study compared with the levels before the study in Group A (1.6 ± 0.9 → 2.0 ± 1.1 g; P< 0.01), while it showed no statistically significant changes in Group B or C.
    Yamamoto et al., International Journal of Food Sciences and Nutrition, 2012 · checked 2026-09-18 · independently re-checked
17. k-07 · Position of an expert body · expert body statement
    Constipation, fatigue, muscle weakness, and malaise. (value from the source's table, not a sentence)
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · extracted
18. k-t13 · Observational data · observational analysis within a randomised trial
    Self-controlled case analyses comparing AF incidence during hypokalaemia (p-potassium <3.5 mmol/L) vs in normal range yielded an incidence rate ratio of 2.24 (1.29-3.88). Hypokalaemia was present in 5.1% of days with AF lasting <60 min and 19.1% with AF lasting >240 min.
    Diederichsen et al., European Heart Journal, 2025 · checked 2026-09-17 · independently re-checked
19. k-r2-3 · Observational data · meta-analysis of prospective observational studies
    For blood and urinary potassium levels, higher level of blood potassium increased the risk of all-cause mortality by 23% (RR: 1.23, 95% CI: 1.11-1.36). The association of blood potassium levels with mortality was nonlinear (Pnon-linearit<0.001). However, urinary potassium levels were inversely associated with the risk of all-cause mortality (RR: 0.84, 95% CI: 0.76-0.93).
    Fan et al., Critical Reviews in Food Science and Nutrition, 2024 · checked 2026-09-17 · independently re-checked
20. k-r5-2 · Observational data · meta-analysis of cohort studies
    Hyperkalemia was significantly associated with an approximately 3.0-fold increased risk of all-cause in-hospital mortality (RR:2.78,95CI%:1.92,4.03), 1.8-fold of all-cause short-term mortality (RR:1.80, 95CI%:1.44,2.27), 1.3-fold of all-cause long-term mortality (RR:1.33, 95CI%:1.19,1.48) and 1.2-fold of cardiovascular mortality (RR:1.19, 95CI%:1.04,1.36).
    Fan et al., Nutrition Journal, 2024 · checked 2026-09-17 · independently re-checked
21. k-r5-3 · Observational data · systematic review and meta-analysis of randomised and observational studies
    We found very low-quality evidence that restricted (1,295 mg/d) versus unrestricted (1,570 mg/d) dietary potassium lowered Sk by -0.22 mEq/L (95% confidence interval [CI]: -0.33, -0.10; I2 = 0%). Lower (1,725 mg/d) versus higher (4,558 mg/d) dietary potassium was not significantly associated with disease progression (hazard ratio [HR]: 1.14; 95% CI: 0.77, 1.70; I2 = 57%). Lower (1,670 mg/d), compared with higher (4,414 mg/d) dietary potassium intake was associated with a 40% reduction in mortality hazard (HR: 0.60; 95% CI: 0.40, 0.89; I2 = 56%).
    Morris et al., Journal of Renal Nutrition, 2020 · checked 2026-09-17 · independently re-checked
22. k-05 · Meta-analysis or systematic review · meta-analysis of randomised trials
    A 2017 meta-analysis of 25 randomized controlled trials in 1,163 participants with hypertension found significant reductions in systolic blood pressure (by 4.48 mm Hg) and diastolic blood pressure (by 2.96 mmHg) with potassium supplementation.
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · independently re-checked
23. k-06 · Observational data · meta-analysis of prospective cohorts
    A meta-analysis of 11 prospective cohort studies in 247,510 adults found that a 1,640 mg per day higher potassium intake was associated with a significant 21% lower risk of stroke.
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · independently re-checked
24. k-t1 · Meta-analysis or systematic review · systematic review and meta-analysis of randomised trials
    Calcium and magnesium achieved significant reductions in both SBP and DBP of -1.37/-1.63 mm Hg and -2.79/-1.56 mm Hg, respectively. Vitamin E and potassium only yielded significant reductions in SBP with values of -1.76 mm Hg and -2.10 mm Hg, respectively.
    Behers et al., Nutrients, 2023 · checked 2026-09-17 · independently re-checked
25. k-t3 · Meta-analysis or systematic review · Cochrane systematic review and meta-analysis
    LSSS compared to regular salt probably reduce DBP on average (mean difference (MD) -2.43 mmHg, 95% confidence interval (CI) -3.50 to -1.36; 20,830 participants, 19 RCTs, moderate-certainty evidence) and SBP (MD -4.76 mmHg, 95% CI -6.01 to -3.50; 21,414 participants, 20 RCTs, moderate-certainty evidence) slightly.
    Brand et al., Cochrane Database of Systematic Reviews, 2022 · checked 2026-09-17 · independently re-checked
26. k-t4 · Meta-analysis or systematic review · Cochrane systematic review and meta-analysis
    probably increase blood potassium slightly (MD 0.12 mmol/L, 95% CI 0.07 to 0.18; 784 participants, 6 RCTs, moderate-certainty evidence), compared to regular salt. LSSS may result in little to no difference, on average, in hypertension (AE 17 fewer/1000, 95% CI -58 to 17; 2566 participants, 1 RCT, low-certainty evidence) and hyperkalaemia (AE 4 more/100,000, 95% CI -47 to 121; 22,849 participants, 5 RCTs, moderate-certainty evidence) compared to regular salt.
    Brand et al., Cochrane Database of Systematic Reviews, 2022 · checked 2026-09-17 · independently re-checked
27. k-t5 · Randomised controlled trial(s) · cluster-randomised controlled trial
    The rate of stroke was lower with the salt substitute than with regular salt (29.14 events vs. 33.65 events per 1000 person-years; rate ratio, 0.86; 95% confidence interval [CI], 0.77 to 0.96; P = 0.006), as were the rates of major cardiovascular events (49.09 events vs. 56.29 events per 1000 person-years; rate ratio, 0.87; 95% CI, 0.80 to 0.94; P<0.001) and death (39.28 events vs. 44.61 events per 1000 person-years; rate ratio, 0.88; 95% CI, 0.82 to 0.95; P<0.001).
    Neal et al., New England Journal of Medicine, 2021 · checked 2026-09-17 · independently re-checked
28. k-t6 · Randomised controlled trial(s) · post hoc analysis of a cluster-randomised controlled trial
    Compared with usual salt group (n = 298), the salt substitute group (n = 313) had a lower hypertension incidence (11.7 vs 24.3 per 100 person-years; adjusted HR: 0.60; 95% CI: 0.39 to 0.92; P = 0.02) but did not increase incidence of hypotension episodes (9.0 vs 9.7 per 100 person-years; P = 0.76).
    Yin et al., Journal of the American College of Cardiology, 2024 · checked 2026-09-17 · independently re-checked
29. k-t7 · Meta-analysis or systematic review · systematic review and meta-analysis of clinical trials
    Potassium supplementation caused a significant decrease in systolic BP (mean difference [95% CI] -3.69 mmHg [-4.91, -2.46], P < 0.001) and increase in serum potassium (+0.37 [0.23, 0.52] mmol/l, P < 0.001). There was an increase in plasma aldosterone (standardized difference 0.426 [0.299, 0.553], P < 0.001) but not in plasma renin activity.
    Ahmad et al., Journal of Hypertension, 2024 · checked 2026-09-17 · independently re-checked
30. k-r5-5 · Meta-analysis or systematic review · meta-analysis of randomised controlled trials
    In this population, salt substitute may reduce risk for all-cause mortality (6 RCTs; 27 710 participants; rate ratio [RR], 0.88 [95% CI, 0.82 to 0.93]; low certainty) and cardiovascular mortality (4 RCTs; 25 050 participants; RR, 0.83 [CI, 0.73 to 0.95]; low certainty). Salt substitute may result in a slight reduction in MACE (3 RCTs; 23 215 participants; RR, 0.85 [CI, 0.71 to 1.00]; very low certainty), with very low-certainty evidence of serious adverse events (6 RCTs; 27 995 participants; risk ratio, 1.04 [CI, 0.87 to 1.25]).
    Greenwood et al., Annals of Internal Medicine, 2024 · checked 2026-09-17 · independently re-checked
31. k-e3-04 · Randomised controlled trial(s) · randomised crossover controlled feeding trial
    Absolute potassium retention was significantly higher during the supplement (28 ± 4.6 mmol/d or 1093 ± 178 mg/d) compared to the control (14.5 ± 3.2 mmol/d or 568 ± 125 mg/d; p < 0.0001), potato (20.6 ± 3.9 mmol/d or 807 ± 154 mg/d; p = 0.04), and FF (21.2 ± 3.5 mmol/d or 829 ± 138 mg/d; p = 0.07) study phases.
    Stone et al., Nutrients, 2021 · checked 2026-09-18 · independently re-checked
32. k-e3-13 · Position of an expert body · guideline research gaps statement
    the effectiveness and safety of LSSS on a participant population that is representative of the general population such as normotensive people and people without history of CVD ... the effectiveness and safety of LSSS in children and pregnant women
    Use of lower-sodium salt substitutes: WHO guideline, World Health Organization, 2025 (CC BY-NC-SA 3.0 IGO) · checked 2026-09-18 · independently re-checked
33. k-e3-14 · Meta-analysis or systematic review · dose-response meta-analysis of randomised controlled trials
    For instance, for a 50 mmol/day increase in 24-h urinary potassium excretion the hypotensive effect was estimated at –5.3 mmHg in SBP in subjects with hypertension compared with only –0.5 mmHg in subjects without hypertension
    Clinical Kidney Journal, 2025 · checked 2026-09-18 · independently re-checked
34. k-e3-15 · Meta-analysis or systematic review · dose-response meta-analysis of randomised controlled trials
    The studies involved a total of 684 patients, 342 in the intervention group (potassium intake modification) and 342 in the control group (unchanged potassium intake). ... Intervention durations ranged from 1 to 6 weeks.
    Clinical Kidney Journal, 2025 · checked 2026-09-18 · independently re-checked
35. k-08 · Position of an expert body · expert body statement
    In people with impaired urinary potassium excretion due to chronic kidney disease or the use of certain medications...even dietary potassium intakes below the AI can cause hyperkalemia.
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · extracted
36. k-r5-1 · Observational data · individual participant data meta-analysis of prospective cohorts
    The risk relationship between potassium levels and adverse outcomes was U-shaped, with the lowest risk at serum potassium of 4-4.5 mmol/L. Compared with a reference of 4.2 mmol/L, the adjusted hazard ratio for all-cause mortality was 1.22 [95% confidence interval (CI) 1.15-1.29] at 5.5 mmol/L and 1.49 (95% CI 1.26-1.76) at 3.0 mmol/L.
    Kovesdy et al., European Heart Journal, 2018 · checked 2026-09-17 · independently re-checked
37. k-r5-4 · Observational data · systematic review and meta-analysis of observational reports
    The pooled prevalence of potassium value &gt;5.0 mmol/L, &gt;5.5 mmol/L and &gt;6.0 mmol/L or symptomatic, was, respectively, 22%, 10% and 0.2%. The analysis disclosed that the risk of trimethoprim-associated hyperkalaemia is dose-related and enhanced by drugs with known hyperkalaemic potential including potassium-sparing diuretics, renin-angiotensin-aldosterone system inhibitors, β-blockers and non-steroidal anti-inflammatory agents.
    Faré et al., Journal of Antimicrobial Chemotherapy, 2022 · checked 2026-09-17 · independently re-checked
38. k-e3-10 · Position of an expert body · systematic review within a guideline
    The systematic review found no meaningful increase in hyperkalaemia with LSSS when compared to regular salt, with little or no difference in effect for this important safety outcome at maximal follow-up of 5 years.
    Use of lower-sodium salt substitutes: WHO guideline, World Health Organization, 2025 (CC BY-NC-SA 3.0 IGO) · checked 2026-09-18 · independently re-checked
39. k-e3-16 · Position of an expert body · agency fact sheet
    Many dietary supplement manufacturers and distributors limit the amount of potassium in their products to 99 mg (which is only about 2% of the DV) because of two concerns related to potassium-containing drugs.
    Potassium, Fact Sheet for Health Professionals, NIH Office of Dietary Supplements · checked 2026-09-18 · independently re-checked
40. k-e3-17 · Position of an expert body · agency fact sheet
    In addition, there is no evidence that high intakes of potassium cause hyperkalemia in adults with normal kidney function or other adverse effects. Therefore, the committee did not set a UL for potassium.
    Potassium, Fact Sheet for Health Professionals, NIH Office of Dietary Supplements · checked 2026-09-18 · independently re-checked
41. k-09 · Position of an expert body · expert body statement
    These medications reduce urinary potassium excretion, which can lead to hyperkalemia.
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Potassium · checked 2026-09-15 · extracted
42. k-t8 · Meta-analysis or systematic review · network meta-analysis of randomised controlled trials
    Two hundred and seventy-six double-blind RCTs involving 58 807 participants (mean age: 55 years; 45% women) were included. All treatment groups were more effective than placebo in lowering BP, with mean differences (MDs) of change from baseline ranging from -7.66 mmHg [95% credible interval (95% CrI), -8.53 to -6.79] for T- to -12.77 mmHg (95% CrI, -15.22 to -10.31) for T+PS-. ... Compared with placebo, all treatments (except T-PS-) were associated with more potassium reduction and T+ compared with all other treatments and T- when compared with T-PS-.
    Cárdenas-Rodríguez et al., Journal of Hypertension, 2023 · checked 2026-09-17 · independently re-checked
43. k-t9 · Meta-analysis or systematic review · model-based meta-analysis of randomised controlled trials
    After a typical follow-up of 8 weeks, fixed-dose combinations of ARB with a high dose (25 mg) of HCTZ were associated with a higher hypokalemia risk difference (RD) from placebo (e.g.,Valsartan + HCTZ: 2.52%[95%CIs:1.17, 4.38%]). However, when ARB was combined with a lower, 12.5 mg dose of HCTZ, hypokalemia RD from placebo was not significant (Valsartan + HCTZ: -0.03%[-0.80, 0.71%]). ARB monotherapy raised hyperkalemia RD from placebo (1.3%[0.3, 3.6%]).
    Schindler et al., Journal of Human Hypertension, 2022 · checked 2026-09-17 · independently re-checked
44. k-t10 · Randomised controlled trial(s) · randomised controlled trial
    Serum K+ level in IND/KCl group decreased from 4.27 ± 0.36 to 3.89 ± 0.28 mmol/L (P < 0.001) ... However, the difference value in UA in IND/KCl group was significantly higher than that in IND group (0.066 (95% confidence interval (CI): 0.041-0.090)  mmol/L vs. 0.029 (95% CI: 0.006-0.058) mmol/L, P < 0.05).
    Zhang et al., Journal of Human Hypertension, 2018 · checked 2026-09-17 · independently re-checked
45. k-t11 · Randomised controlled trial(s) · post hoc analysis of a randomised controlled trial
    In a multivariate analysis, the following variables were independently associated with the development of hypokalemia: baseline K+ values (OR per 0.1 units, 0.82 [95% CI, 0.76-0.87]; P<0.001), treatment with HCTZ (OR, 4.90 [95% CI, 2.50-9.90]; P<0.001), and treatment with a mineralocorticoid receptor antagonist at baseline (OR, 0.42 [95% CI, 0.20-0.84]; P=0.017).
    Cobo-Marcos et al., Circulation: Heart Failure, 2025 · checked 2026-09-17 · independently re-checked
46. k-t12 · Meta-analysis or systematic review · meta-analysis of individual participant data from randomised controlled trials
    SGLT2 inhibitors reduced the risk of serious hyperkalemia (hazard ratio, 0.84 [95% CI, 0.76-0.93]), an effect consistent across studies (Pheterogeneity=0.71). The incidence of investigator-reported hyperkalemia was also lower with SGLT2 inhibitors (hazard ratio, 0.80 [95% CI, 0.68-0.93]; Pheterogeneity=0.21).
    Neuen et al., Circulation, 2022 · checked 2026-09-17 · independently re-checked
47. k-e3-11 · Position of an expert body · guideline appraisal of evidence generalisability
    All included trials specifically excluded participants in whom an increased intake of potassium could cause harm - for example, people with kidney disease, impaired renal function or those using potassium-sparing medications.
    Use of lower-sodium salt substitutes: WHO guideline, World Health Organization, 2025 (CC BY-NC-SA 3.0 IGO) · checked 2026-09-18 · independently re-checked
48. k-e3-12 · Position of an expert body · guideline implementation guidance
    the use of LSSS should be implemented in settings with adequate access to health care, where kidney disease would not go undiagnosed for a long time
    Use of lower-sodium salt substitutes: WHO guideline, World Health Organization, 2025 (CC BY-NC-SA 3.0 IGO) · checked 2026-09-18 · independently re-checked