Three minerals do most of the work: sodium sets your blood pressure and fluid balance, potassium pulls it back down, magnesium helps your heart, your sleep, and a few hundred other things. The Western default — too much sodium, not enough of the other two — is one of the few dietary patterns where the cardiovascular evidence is hard. Hypertensives get the biggest win. Endurance athletes get a narrower but real one. Most people get something in between, and the cramps story is mostly a myth.
Inside every cell, a small protein pump traded three sodium ions out for two potassium ions in, roughly a billion times a second, while you read that sentence. That pump — the Na/K-ATPase — is doing about a third of the work of your resting metabolism. It sets the voltage on every nerve, every heartbeat, every muscle twitch. Sodium and potassium are the two ions that voltage runs on. Magnesium is the cofactor that lets the engine room run at all — every molecule of ATP your body uses is functionally bound to a magnesium ion.
Three minerals, three jobs:
- Sodium lives outside your cells, holds water in the bloodstream, and (when there's too much of it) raises blood pressure by raising the volume in your pipes.
- Potassium lives inside your cells, dilates your arteries, and tells your kidneys to dump extra sodium. It is sodium's natural counterweight.
- Magnesium is a cofactor for about 600 enzymes, gently blocks calcium from over-firing in your blood vessels (so they relax), and quiets the brain's main "go" receptor at night (so you sleep) Schwalfenberg 2017.
The Western pattern jams the first one up and starves the other two. That's the whole story in one sentence; the rest of this entry is the texture.
What the trials actually show
The blood-pressure case is the cleanest. In the DASH-Sodium trial, people with high blood pressure who moved from a typical American sodium intake to a low one — while eating a fruit, vegetable, and dairy-heavy pattern — dropped their top number by about 7 points in under a month Sacks 2001. Seven points is about what a starting dose of a real blood-pressure drug does. A Cochrane review pooling 185 trials lands in the same place for hypertensives, with a smaller effect (~1.3 mmHg) for people whose pressure was already normal Graudal 2017.
Potassium runs the opposite direction with similar force: three and a half points off the top number in hypertensives who raised their potassium intake, plus a roughly 24% lower stroke risk in the long run Aburto 2013b. The dose-response flattens out around 3.5 grams a day — beyond that, more doesn't buy more Filippini 2020.
The most convincing trial is the one that combined both ideas in one move.
Magnesium's case is real but narrower. A 34-trial meta-analysis found that about 370 mg a day for three months dropped systolic pressure by another two points, with bigger effects in people who were deficient to start Zhang 2016. Stack it on top of the sodium-and-potassium move and the effects roughly add up. About half of US adults don't hit the basic magnesium intake target, so the deficient-subgroup result applies to a lot of people Rosanoff 2012.
Two important counterweights, because the field is not unanimous. The PURE study, which tracked salt intake and cardiovascular events in 95,000 people across 18 countries, found a J-shape: both very high and very low sodium intake associated with worse outcomes, with the sweet spot around 3 to 5 grams a day — somewhat above the strict WHO target Mente 2018. And the SODIUM-HF trial in heart failure patients found that pushing sodium below the moderate band gave no extra survival benefit at all Ezekowitz 2022. The cleanest reading: get from the very-high band down into the moderate band; chasing zero is not where the evidence is.
If you keep eating the default
The typical Western day delivers somewhere north of 3.5 grams of sodium, around 2 grams of potassium, and 250 milligrams of magnesium — too much of one, not enough of the other two Cogswell 2016, Rosanoff 2012. The damage doesn't show up tomorrow. It shows up in your blood-pressure cuff at the next physical, a few points higher than last time, and a few points higher than that the year after.
Years in: the doctor mentions the word borderline. Then drops it. A decade in, you're on a small daily pill, and you don't think much about it because everyone you know is on one too. Two decades in, the artery wall has been remodeling around the higher pressure long enough that the consequences start landing — the stroke that takes a friend's father out, the heart-attack scare that puts your colleague on disability. About half of US adults are hypertensive by their late fifties; the sodium-and-potassium pattern is one of the modifiable drivers most of them never adjusted Whelton 2018.
There's a quieter parallel story for magnesium. Sleep that takes a bit longer to come, year after year. The leg cramp that wakes you at 4 a.m. and you blame on the gym. A migraine pattern that doesn't quite respond to anything. None of these are diagnostic by themselves, but at the population level, the half of adults who are below the magnesium target show small elevations in inflammatory markers, insulin resistance, and migraine frequency Rosanoff 2012, Schwalfenberg 2017. It's the slow grind, not a crisis.
What to actually do
The targets, in plain numbers. These are the WHO and US guideline figures, and they line up with where the trial evidence pulls:
Three practical moves, in order of how much they matter:
Cook more meals from raw ingredients. Over 70% of the sodium in a typical American diet comes from packaged and restaurant food, not what you sprinkle on it Cogswell 2016. Bread, deli meat, cheese, soup, sauces, anything in a box. The saltshaker is the symbol; the box is the problem. Cooking from raw, even with a confident hand at the salt, lands most people near 2–3 grams sodium a day on its own.
Swap your table salt for a potassium-enriched one. LoSalt, NoSalt, Morton Salt Substitute — they're about half regular salt and half potassium chloride. The SSaSS trial used a 75/25 version Neal 2021. Cheap, invisible at the dinner table once you adjust, and it nudges both the sodium and potassium numbers in the right direction at once. Worth a clinician check first if you take a blood-pressure medication that already raises potassium (ACE inhibitors, ARBs, spironolactone) or have kidney trouble.
Get your potassium from food, not pills. Pharmacy potassium tablets are capped at 99 milligrams each by the FDA because rapid-release potassium can burn the stomach lining and disturb heart rhythm. A baked potato is about 900 milligrams. A cup of cooked spinach is 840. Six ounces of plain yogurt is 400. The math says food.
For workouts: drink to thirst for sessions under an hour. For longer or hotter sessions, mix about 500 milligrams of sodium per half-litre of water — a quarter-teaspoon of table salt in a bottle, optionally with a splash of fruit juice for taste and potassium McDermott 2017. Commercial sports drinks and electrolyte sticks do the same job at a higher markup; if you like them, fine. The same ratio principle — not plain water — is what rehydrates you after the fluid loss of travel diarrhea or a stomach bug, which is why oral rehydration sachets are a travel-kit staple.
When the rebalancing is dangerous
The widely-repeated stuff that isn't true
"Cramps mean you're low on electrolytes." The gym-floor consensus that a leg cramp at mile twenty is a sodium or magnesium deficit doesn't survive contact with the data. Field studies on marathon runners and Ironman finishers found cramping and non-cramping athletes had statistically identical post-race sodium and magnesium levels Schwellnus 2009. The better explanation is altered neuromuscular control — fatigued spinal nerves firing on their own. Pickle juice works in about ninety seconds, which is too fast for any mineral to absorb; the cramp relief is a reflex from the back of the throat, not a refill. For older adults' leg cramps, a 2020 Cochrane review of 11 trials found magnesium didn't beat placebo Garrison 2020. If you cramp a lot, stretching, training adaptation, and pace management are the higher-evidence answers.
"Drink ahead of your thirst." The sports-drink industry sold this for decades. The actual marathon-medicine consensus, after enough runners died from over-drinking plain water, is the opposite: drink to thirst. Between 13% and 18% of marathon finishers cross the line with blood sodium below the normal range from sheer water dilution; the severe cases get brain swelling and sometimes die from it Hew-Butler 2015. Thirst is a reasonable signal. Trust it.
"Less salt is always better." Cleaner version: less salt is better when you're starting from too much. Pushing below the moderate band — into the 1-gram-a-day territory — has not delivered extra benefit in trials and may activate the body's stress-and-water-retention system in unhelpful ways Mente 2018, Ezekowitz 2022. The big lever is moving from the high band to the moderate band, not chasing zero.
"Magnesium oxide is fine — they're all the same." They're not. The cheap white tablets on the pharmacy shelf are usually magnesium oxide, which your gut absorbs at about 4 to 10 percent. Magnesium glycinate, citrate, malate, and threonate absorb at three to four times that rate, and the glycinate form doesn't pull water into your gut the way citrate can at high doses Schwalfenberg 2017. Read the label; the form matters more than the milligram count.
Where this goes wrong in practice
You cut salt at the table and your sodium barely moves. The shaker accounts for a small fraction of what you eat. The pizza on Friday, the deli sandwich on Tuesday, the canned soup, the bottled marinade — those are where the grams are. People who feel virtuous about the saltshaker and then can't figure out why their pressure didn't drop are looking at the wrong lever Cogswell 2016.
You cut too hard and feel awful. Below about a gram and a half a day, especially if you're also on a water pill or a low-carb diet that's already wringing sodium out of you, you can get fatigue, lightheadedness, and a low-grade headache that doesn't quit. Most people who report "I tried a low-salt diet and it ruined me" were sitting in this range without realizing it. Aim for the moderate band, not the floor.
The magnesium supplement does nothing. Three common reasons: it's oxide and you're absorbing almost none of it; the dose is too low (a lot of products are 50 milligrams a capsule); you take it in the morning and expect the sleep benefit at night. Switch to glycinate or citrate, get to 200 to 400 milligrams, take it after dinner.
Endurance fueling on plain water in heat. A four-hour bike ride, a hot marathon, a long hike — at high sweat rates, plain water dilutes your blood sodium faster than your kidneys can compensate. The person you'll worry about is the one who drank a litre an hour and feels increasingly confused at hour three. Sodium in the bottle is the fix; the dose is roughly a quarter-teaspoon of salt per half-litre McDermott 2017.
Who gets the biggest payoff
The rebalancing is universal in principle. The size of the win is not.
Older adults. Three things compound here: your thirst signal blunts with age (so dehydration sneaks up), your kidneys lose magnesium faster, and the loop diuretics that get prescribed for blood pressure and heart conditions drain magnesium and potassium together Hooper 2020. The magnesium-for-sleep effect is largest in this age band — about a quarter-hour faster to fall asleep, in trials of insomniac older adults on 500 milligrams a day for two months Abbasi 2012, Mah 2021. The blood-pressure payoff is also larger, because age and salt sensitivity track together Whelton 2018.
Anyone already hypertensive. The blood-pressure response to both sodium reduction and potassium loading runs about two to three times larger than in normotensives Graudal 2017. Same protocol; much bigger result — and the ratio fix is the dietary cornerstone of the first ninety days after a hypertension diagnosis.
Endurance athletes and heavy sweaters. If you can see white salt streaks on a dark shirt after a workout, your sweat sodium concentration is genuinely high (sweat sodium varies fivefold across people), and you need genuinely more sodium during long efforts. This is the one population where the commercial electrolyte-drink ecosystem matches a real physiological need McDermott 2017.
Low-carb and keto eaters. Keeping insulin low tells the kidneys to dump sodium. You actually do need more salt on these diets — most of "keto flu" is sodium deficit. Three to five grams a day is reasonable here; this is the one place the "salt is harmless" influencer message lands closer to the data than the standard guidance does.
Sedentary normotensives at moderate intake. If your pressure is normal, you cook most meals, and your sodium is already in the moderate band, the marginal blood-pressure case is small. The marginal potassium-and-magnesium case is bigger — most people in this group are still short on both.
What changes if you actually do it
Week one. If you've started cooking from raw and your sodium intake was high, you may notice the food tasting flatter for a few days, then sharpening again as your taste adjusts. Roughly a quarter of people pee more in the first few days — that's the extra plasma volume coming off.
Weeks two to four. If you had high blood pressure, the cuff at the next physical reads five to seven points lower on the top number Sacks 2001. If you were short on magnesium and slept badly, you start falling asleep a quarter-hour faster, and that 4 a.m. leg cramp stops happening Abbasi 2012. Energy on long workouts feels steadier — less of the late-session crash, less GI distress.
Months in. The doctor talks about reducing a dose, or holding off on starting one. Your partner stops mentioning that you snore-and-startle in the early morning. The migraine pattern, if you had one, eases a step — magnesium's case here is mechanistic and not airtight, but a real fraction of migraine sufferers respond Maier 2020. None of this is dramatic. It's the absence of a slow worsening.
Years in. This is where the trial-level evidence pays. In the SSaSS population over five years, the salt-substitute arm had 14% fewer strokes, 13% fewer major cardiovascular events, and 12% fewer deaths from any cause than the regular-salt arm Neal 2021. That's one ingredient swap, in cooking, for five years. The reader who makes the change at 40 and holds it across the decades buys themselves a meaningful shift in their cardiovascular trajectory — the kind of shift that does not feel like anything until the day a friend is unlucky and you aren't.
Related, worth a look
- Calcium and bone — the fourth canonical electrolyte; covered in its own entry because the dominant story is skeletal, not balance.
- Water intake — how much plain water to drink; the volume question, separate from the mineral question.
- The DASH dietary pattern — the whole-diet version of this entry's protocol.
- Blood-pressure monitoring at home — how to actually measure whether any of this is working for you.
- Endurance fueling — the deeper protocol for athletes doing multi-hour efforts in heat.
- Sleep hygiene and CBT-I — the higher-evidence path for sleep problems where magnesium doesn't move the needle.
- — Fixing the sodium-potassium ratio is one of the few diet moves with hard blood-pressure evidence.
- — Electrolytes work alongside plain hydration; the ratio matters most once your daily water intake is actually adequate.
- — The electrolyte most diets run short on is potassium; its ratio to sodium is the whole game.
- — For travel-diarrhea fluid loss it's the salt ratio, not plain water, that rehydrates — which is why oral rehydration packets are kit staples.
- — Magnesium is the third mineral here — heart rhythm, sleep, and a few hundred other jobs.
- — Potassium salt substitutes are a concrete way to shift the sodium-potassium ratio in your favour.
- — Fixing the sodium-potassium ratio is the big lever; which salt you reach for is the small one, and it also decides whether you get enough iodine.
- — Remineralizing RO water is one small way the minerals that matter get back into your day.
Substance and claimed effects
The entry covers three dietary electrolytes — sodium, potassium, and magnesium — and the balance between them through whole-food intake and supplementation. The cation trio sets extracellular and intracellular ionic gradients, drives action potentials in nerve and cardiac muscle, governs renal water handling, and modulates vascular tone. Claimed consequences span six dimensions named in the brief: hydration (sodium-driven plasma volume retention; oral rehydration), blood pressure (sodium-raising / potassium-lowering / magnesium-modestly-lowering), muscle cramps (longstanding folk attribution to sodium or magnesium loss, evidence mixed), exercise performance (volume + electrolyte replacement during prolonged effort), sleep (magnesium's GABAergic and NMDA-blocking effects in deficient or older adults), and cardiac rhythm (potassium and magnesium as direct determinants of myocardial repolarization; hypokalemia and hypomagnesemia as arrhythmia triggers). Calcium is a fourth canonical electrolyte but is excluded — its dominant story is bone, not balance, and warrants a separate entry. Chloride moves passively with sodium and is not separately scored.
Evidence by addressing question
mechanism
Sodium is the principal extracellular cation; serum concentration is held at 135–145 mmol/L by tight osmoregulation. Retained sodium pulls water into the vascular compartment, raising plasma volume and — via baroreflex and renin-angiotensin-aldosterone (RAAS) feedback — arterial pressure NASEM 2019. Potassium is the principal intracellular cation; the Na/K-ATPase pump trades three sodium ions out for two potassium ions in across every cell membrane, establishing the resting membrane potential. Dietary potassium raises kallikrein-kinin activity, dilates resistance arterioles, and natriureses sodium — three independent mechanisms producing the well-documented blood-pressure drop Aburto 2013b. Magnesium is a cofactor for ~600 enzymatic reactions including ATP-Mg binding (every ATP molecule is functionally Mg-ATP), serves as a natural calcium-channel antagonist, blocks NMDA receptors at the resting state, and supports parathyroid hormone secretion Schwalfenberg 2017. Magnesium's vasodilatory effect runs through smooth-muscle calcium displacement; its sleep effect runs through GABA-A receptor agonism and NMDA antagonism Maier 2020.
Cardiac rhythm depends on tightly bounded potassium and magnesium. Repolarization in ventricular myocytes is potassium-current-driven; hypokalemia prolongs QT and predisposes to torsades de pointes, while hypomagnesemia inhibits the inward-rectifier K-channels and likewise prolongs repolarization Schwalfenberg 2017. The two deficiencies tend to co-occur because renal magnesium reabsorption requires intracellular potassium.
evidence
Sodium and blood pressure. DASH-Sodium (n=412, 30-day crossover) showed a stepwise systolic-BP drop of 7.1 mmHg in hypertensives moving from high (~3.5 g sodium/day) to low (~1.5 g/day) intake while eating the DASH diet Sacks et al. 2001. A Cochrane meta-analysis of 185 RCTs found a sodium-restriction effect of ~5.5/2.9 mmHg in hypertensives and ~1.3/0.8 mmHg in normotensives Graudal 2017. The observational picture is messier: the PURE cohort (n=95,767 across 18 countries) found a J-shaped curve where both very low (<3 g/day) and very high (>7 g/day) sodium excretion associated with elevated cardiovascular events, with the lowest risk in the 3–5 g/day band Mente 2018. The J-curve is contested (reverse causation, spot-urine sampling artefact); the Aburto BMJ meta-analysis of 36 trials supports a monotonic BP benefit and modest cardiovascular benefit of reduced sodium Aburto 2013a.
Potassium and blood pressure / stroke. Aburto's parallel BMJ meta-analysis of 22 RCTs found increased potassium intake reduced systolic BP by 3.5 mmHg in hypertensives and stroke risk by ~24% Aburto 2013b. Filippini's 2020 dose-response meta-analysis identified ~3.5 g/day as the inflection where additional potassium stops moving the needle Filippini 2020. SSaSS (n=20,995 in rural China; salt substitute replacing 25% NaCl with KCl) is the cleanest large trial: it cut stroke incidence by 14%, major cardiovascular events by 13%, and all-cause mortality by 12% over five years Neal 2021. The intervention is genuinely additive — half the effect comes from less sodium, half from more potassium.
Magnesium and blood pressure. A 34-RCT meta-analysis (n=2,028) found that ~370 mg/day of supplemental magnesium for ≥3 months reduced systolic BP by 2.0 mmHg and diastolic by 1.8 mmHg, with larger effects in deficient and hypertensive subgroups Zhang 2016. Smaller than sodium or potassium, but real and additive.
Magnesium for sleep. Abbasi's RCT in 46 older adults with insomnia: 500 mg/day magnesium for 8 weeks increased sleep time by ~17 minutes, improved sleep-onset latency by ~17 minutes, and raised serum melatonin Abbasi 2012. A 2021 BMC systematic review pooled three RCTs (n=151) in older adults: modest reduction in sleep-onset latency (-17 min) but low certainty given small studies Mah 2021. The signal is real but narrow — older adults, baseline insomnia, low magnesium status.
Cramps. The mechanism story (sodium loss → cramps) is folk-medicine and gym-floor consensus; the evidence base is weaker than the consensus implies. Schwellnus's review and field studies of marathon runners and Ironman athletes found cramping runners and non-cramping runners had statistically indistinguishable post-race sodium and magnesium levels, supporting an altered-neuromuscular-control theory (fatigued spinal motor neurons firing unprompted) over depletion Schwellnus 2009. The 2020 Cochrane meta-analysis of 11 RCTs (n=735) for magnesium in skeletal-muscle cramps in older adults and pregnant women found no clinically meaningful benefit over placebo Garrison 2020. Pickle-juice case reports show cramp relief within ~85 seconds — too fast for sodium absorption, suggesting an oropharyngeal-reflex mechanism, not electrolyte replacement.
Exercise / hydration. The NATA position statement endorses electrolyte replacement during exercise lasting over 60 minutes or in heavy sweating, with sodium 300–700 mg/L as the target concentration in fluids McDermott 2017. Below that intensity, water alone is sufficient. The opposing edge of the field, the Hew-Butler consensus on exercise-associated hyponatremia, warns that overdrinking plain water in endurance events is the dominant risk: 13–18% of marathoners finish with serum sodium below 135 mmol/L; severe cases (<125 mmol/L) cause cerebral edema and death Hew-Butler 2015. Drink to thirst is the consensus recommendation; sodium intake protects against dilutional hyponatremia at high sweat rates.
Heart failure — the SODIUM-HF surprise. The 2022 SODIUM-HF trial (n=806) tested aggressive sodium restriction (<1.5 g/day) vs. usual care (~2.2 g/day) in stable heart failure. No difference in cardiovascular hospitalization or mortality at 12 months Ezekowitz 2022. Modest QoL improvements only. The trial pushed cardiology away from severe sodium restriction in HF outside of decompensation.
protocol
Targets, sourced from guidelines and the dose-response evidence above:
- Sodium: 1.5–2.3 g/day for adults with hypertension or elevated CV risk (~3.8–5.8 g salt); 2.3 g/day sodium cap for healthy adults per WHO and 2017 ACC/AHA WHO 2012a, ACC/AHA 2018. NASEM's 2019 chronic-disease-risk reduction (CDRR) intake is 2.3 g/day NASEM 2019. US mean intake is 3.6 g/day, above all of these Cogswell 2016.
- Potassium: WHO recommends ≥3.5 g/day from food WHO 2012b; NASEM Adequate Intake is 3.4 g/day men and 2.6 g/day women NASEM 2019. US mean intake is ~2.6 g/day (men) and ~2.0 g/day (women) — most adults fall short.
- Magnesium: RDA 400–420 mg/day men, 310–320 mg/day women. Supplemental dose for hypertension or insomnia in deficient subjects: 200–400 mg/day elemental as glycinate, citrate, or threonate (better tolerated than oxide, which is poorly absorbed and laxative-like at dose). ~50% of US adults are below the EAR Rosanoff 2012.
- Hydration during exercise: drink to thirst for sessions <60 min; add 300–700 mg sodium/L for sessions >60 min in heat or for "salty sweaters" (visible salt streaks on clothing) McDermott 2017.
The food-vs-supplement split: sodium needs no supplementation in modern diets (food supplies it). Potassium is best from food (fruit, vegetables, legumes, dairy, fish) — supplements over 99 mg per tablet are FDA-restricted because rapid potassium absorption can cause gastric ulcers and arrhythmia. Magnesium can be safely supplemented; food sources (leafy greens, nuts, seeds, whole grains) leave most diets short of the RDA.
contraindications
Kidney disease (CKD stage 3+): impaired potassium excretion makes high-potassium loads dangerous. Hyperkalemia >6 mmol/L is a medical emergency. Salt substitutes containing KCl and any potassium supplement should be cleared with a clinician Neal 2021 explicitly excluded CKD.
Heart failure on RAAS inhibitors / spironolactone / eplerenone: these drugs raise serum potassium independently; added KCl can tip into hyperkalemia.
Adrenal insufficiency / Addison's: impaired aldosterone makes sodium losses dangerous and potassium retention high.
Cardiac arrhythmias on digoxin: hypokalemia and hypomagnesemia potentiate digoxin toxicity; supplementation should be clinician-monitored.
Magnesium and renal failure: renal excretion is the only meaningful elimination route. CKD reduces clearance; high-dose supplementation can cause hypermagnesemia (weakness, hypotension, cardiac arrest at extremes).
Drug interactions: magnesium chelates fluoroquinolone and tetracycline antibiotics, bisphosphonates, and levothyroxine — separate dosing by 2+ hours. Loop and thiazide diuretics deplete all three electrolytes.
misconceptions
"Cramps mean you're low on electrolytes." The evidence does not support this Schwellnus 2009, Garrison 2020. Most exercise-associated cramps are neuromuscular, not depletion. Sports-drink marketing built the depletion myth.
"Drink to stay ahead of thirst." Overdrinking water during endurance events is the dominant fluid-related cause of marathon death — exercise-associated hyponatremia Hew-Butler 2015. Drink to thirst is the safer rule.
"Low-salt diet is universally better." The PURE J-curve and SODIUM-HF complicate this Mente 2018, Ezekowitz 2022. For normotensives at moderate intake (~3–5 g/day sodium), further restriction provides minimal benefit. The big lever is moving from very high intake down to the moderate band, not chasing 1.5 g.
"Sodium drives BP — leave potassium alone." Population BP impact of raising potassium is comparable to lowering sodium, and the two effects compound — that's the SSaSS finding Neal 2021. The Western pattern is sodium-high and potassium-low; both axes matter.
"Magnesium oxide is fine." Bioavailability of MgO is ~4–10%, vs. 30–40% for magnesium citrate, glycinate, malate, and threonate. The cheap MgO supplements that flood the market often deliver almost no absorbed magnesium and cause loose stool at any meaningful dose Schwalfenberg 2017.
failure-modes
Sodium restriction fails in two ways: (1) people swap restaurant meals for restaurant salads and still hit 4 g/day because >70% of US sodium intake comes from processed and restaurant food, not the saltshaker Cogswell 2016; (2) people cut sodium too hard, get fatigue, lightheadedness, and headache (especially when also on a diuretic), then conclude electrolytes don't matter.
Potassium "supplementation" fails because OTC supplements are capped at 99 mg/tablet (a banana is ~400 mg); to meaningfully raise potassium intake from supplements requires multiple tablets per meal, which is unrealistic. Real strategy: food (legumes, leafy greens, potatoes, fish, dairy) plus potassium-enriched salt substitutes.
Magnesium supplementation fails through (1) wrong form (oxide), (2) wrong dose (too low — many products are 50 mg/cap), (3) wrong timing if expecting sleep benefit (take in the evening), (4) GI intolerance from citrate/citrate-magnesium pulling water into the gut (use glycinate instead).
Endurance fueling fails when athletes either over-hydrate with plain water (hyponatremia Hew-Butler 2015) or under-replace sodium on hot multi-hour efforts (cramping, performance drop, GI distress).
stakes
The reader who continues with the Western pattern — sodium 3.5+ g/day, potassium ~2 g/day, magnesium ~250 mg/day — sits in the highest-risk corner of the PURE distribution Mente 2018. Lifetime BP trajectory: a meaningful fraction of the 30% of US adults who will develop hypertension carry the dietary-electrolyte profile as one of the modifiable drivers Whelton 2018. Stroke risk follows: the SSaSS arm without the potassium-rich salt substitute had a 14% higher stroke event rate over five years Neal 2021. At the cellular level, chronic suboptimal magnesium status is associated with insulin resistance, migraine, leg cramps, and elevated CRP — the slow grind of subclinical dysfunction Rosanoff 2012, Schwalfenberg 2017.
payoff
The reader who shifts toward the evidence-based balance — sodium ~2.3 g/day, potassium ~3.5 g/day, magnesium hitting RDA — typically sees BP drop within weeks: ~5–7/3 mmHg in hypertensives on the DASH-Sodium pattern, with onset by week 2 Sacks 2001. Magnesium-deficient subjects starting 300–400 mg/day report sleep-onset improvement within 1–4 weeks Abbasi 2012, Mah 2021. Stroke and CV-event payoff aggregates over years: a 14% relative reduction at the population level when you commit to the swap Neal 2021. Day-to-day energy and exercise recovery often improve subjectively, but the trial data on energy specifically is thin — most of the felt-energy effect is downstream of better BP control, better sleep, and adequate hydration rather than a direct electrolyte stimulant effect.
practicalities
Sodium: don't track the saltshaker; track ultra-processed food intake. Bread, deli meat, soup, sauces, restaurant meals are the dominant contributors. Cooking from raw ingredients with a moderate hand naturally lands near 2–3 g/day Cogswell 2016.
Potassium: easiest single swap is potatoes, beans, leafy greens, and yogurt in regular rotation. A banana is ~400 mg, a baked potato ~900 mg, a cup of cooked spinach ~840 mg, 6 oz yogurt ~400 mg. A potassium-enriched salt substitute (LoSalt, Morton Salt Substitute, NoSalt) is a cheap force multiplier — substitutes typically 50/50 NaCl/KCl. SSaSS used a 75/25 substitute Neal 2021.
Magnesium: 200–400 mg elemental in glycinate, malate, or threonate form, taken in the evening if targeting sleep. Cost ~$10–25/month. Magnesium citrate is a useful budget alternative if GI tolerance is fine.
Exercise drinks: water alone for <60 min sessions. For longer sessions in heat, mix ~500 mg sodium per 500 mL (a quarter-teaspoon table salt in a water bottle, optionally with a splash of fruit juice) — cheaper and more dose-appropriate than commercial sports drinks McDermott 2017. Commercial "electrolyte powders" (LMNT, Liquid IV, etc.) are sodium-heavy by design (~500–1000 mg per stick); fine for athletes and salty sweaters, overkill for sedentary office workers.
audience
Older adults (60+) have blunted thirst, declining renal magnesium reabsorption, and higher loop-diuretic prescription rates. They benefit disproportionately from the magnesium-for-sleep and potassium-for-BP cases, and are at elevated risk for dehydration Hooper 2020, Mah 2021.
Endurance athletes and heavy sweaters need genuinely more sodium during exercise — the only population where the commercial-electrolyte-drink ecosystem matches a real physiological need McDermott 2017.
Hypertensives have ~2–3× the BP response to sodium reduction and potassium loading vs. normotensives Graudal 2017. Higher leverage; same protocol.
Pregnancy: sodium needs are not reduced. Heavy sodium restriction in pregnancy is contraindicated. Magnesium needs rise modestly.
Vegan / very-low-dairy diets: potassium is usually higher (plant-rich); sodium may be lower; magnesium often adequate if leafy greens and legumes are regular.
alternatives
The DASH dietary pattern as a whole captures most of the electrolyte-rebalancing benefit without explicit tracking — DASH naturally delivers ~4.5 g/day potassium, ~500 mg/day magnesium, and trim sodium when followed Sacks 2001, Stamler 2018. For hypertensives unable to modify diet, antihypertensive medication is the alternative (and is often added to the dietary intervention). For magnesium-related sleep issues, conventional sleep hygiene, CBT-I, and (in chronic insomnia) sleep-specialist evaluation are the higher-evidence path.
history
Modern sodium-restriction guidance emerged from INTERSALT and INTERMAP epidemiology and was operationalized in the 1990s by the DASH and DASH-Sodium trials Sacks 2001, Stamler 2018. The "you need electrolytes" sports-drink narrative dates from 1965 Gatorade and was retrofitted with thin science. The cramp-prevention story has never had a strong evidence base but has remained commercially load-bearing for the sports-nutrition industry. The 2010s wave of "salt is fine" contrarianism (Mente, low-carb-influencer communities, the keto-flu narrative) is real and partially correct for normotensives at moderate intake — but the message has been amplified well beyond the data, especially via paleo/keto influencers who conflate "your body needs sodium" (true) with "salt is harmless at any intake" (not true for the half of adults who are salt-sensitive).
out-of-scope
Calcium (own entry), trace minerals (zinc, copper, iodine, selenium — separate supplementation discussion), water intake quantity per se (covered in a hydration entry), DASH diet as a pattern (own entry candidate), specific endurance-fueling protocols (separate entry for athletes), hyperaldosteronism and other rare endocrine causes of electrolyte derangement (medical literature), IV electrolyte replacement (clinical only).
The credibility range
Optimist case
Electrolyte balance is one of the few dietary levers where the mechanism, RCT evidence, and population-scale trial data converge. The SSaSS trial is a hard endpoint cardiovascular RCT — 14% stroke reduction, 12% all-cause mortality reduction, in 21,000 people, from a single dietary swap Neal 2021. The DASH-Sodium effect size in hypertensives rivals a low-dose antihypertensive Sacks 2001. Aburto's meta-analyses on both sodium and potassium independently confirm the BP and stroke effects Aburto 2013a, Aburto 2013b. Magnesium's case is narrower but real: a modest BP effect, a real sleep effect in deficient older adults, and the mechanism (cofactor for 600 enzymes, calcium-channel antagonism) is overwhelming. Adequate magnesium is almost certainly net-beneficial for ~half the US population that is below the EAR Rosanoff 2012.
Skeptic case
The strongest skeptic move is on the BP endpoint vs. hard outcomes. PURE's J-curve for sodium suggests that aggressive restriction below ~3 g/day may not pay off and could harm (compensatory RAAS activation) Mente 2018. SODIUM-HF's null result in heart failure was a major cardiology surprise Ezekowitz 2022. Cochrane Graudal estimates the absolute BP drop in normotensives at ~1.3/0.8 mmHg — clinically near-imperceptible Graudal 2017. The cramp story is essentially myth-driven; the Cochrane magnesium-for-cramps meta-analysis is null Garrison 2020. The magnesium-for-sleep evidence base is three small RCTs in older adults Mah 2021 — not enough to recommend universally. Commercial "electrolyte hydration" products are mostly solving a problem the sedentary office worker doesn't have, and most of the "electrolyte deficiency" symptom narrative on Reddit and Twitter is not grounded in measured deficiency.
Author's call
The optimist case wins for hypertensives, older adults, and anyone in the Western dietary pattern (sodium-high, potassium-low). For normotensives at moderate sodium intake, the BP-restriction case is weaker than guidelines imply, but the potassium-up and magnesium-adequate cases remain solid. Cramps are not an electrolyte problem in most cases — don't promise that fix. Exercise hydration is a real but narrow case: long sessions in heat, not the daily commute. SODIUM-HF tempers the "always less sodium" reflex; SSaSS strengthens the "shift toward potassium" reflex. The entry lands on a balanced rebalancing message, evidence-rated 4 (multiple RCTs, hard endpoints, some heterogeneity), controversy 3 (PURE J-curve and SODIUM-HF are genuine ongoing debates).
Stakeholder and incentive map
- Sports-nutrition industry (Gatorade, Powerade, LMNT, Liquid IV, Nuun): commercial incentive to broaden the "you need electrolytes" message beyond athletes. Sodium-heavy formulations sell on flavor and the cramp-prevention myth.
- Supplement industry — magnesium: real underlying need (half of adults below EAR) but flooded with low-bioavailability MgO products and unsubstantiated claims about anxiety, depression, cognition.
- Salt industry / processed-food lobby: pushback against sodium-reduction guidelines historically vocal (Salt Institute, before its 2019 dissolution). Reformulation toward lower-sodium products has been slow.
- Public-health bodies (WHO, AHA, USDA, NICE): pro sodium reduction, pro potassium increase, broadly aligned. The aggressive 1.5 g/day sodium target is contested even inside this camp.
- Cardiology guidelines bodies (ACC/AHA, ESC): broadly aligned with population sodium reduction; tempered on HF-specific restriction post-SODIUM-HF.
- Keto / paleo / carnivore community: pushes higher sodium intake (some basis: low-carb diets increase renal sodium loss) but often overshoots into "salt is harmless." Magnesium and potassium supplementation are common in this culture and partly evidence-aligned.
- Endurance-sport community: high sodium-replacement custom, partly evidence-based for genuine hot multi-hour effort, partly placebo / cramping-myth.
Population variability
Salt sensitivity is real and bimodal: roughly half of hypertensives and ~25% of normotensives are "salt-sensitive," meaning their BP responds substantially (≥5 mmHg systolic) to sodium change. Salt sensitivity rises with age, kidney function decline, Black ancestry, and obesity Whelton 2018.
Genetic potassium handling: Liddle syndrome, Gitelman syndrome, Bartter syndrome are rare but produce dramatic baseline electrolyte differences.
Magnesium status varies with diet (low in refined-grain-heavy diets), GI absorption (lower in Crohn's, celiac, PPI use), and renal handling (lower with loop diuretics).
Sweat sodium concentration varies ~5-fold across individuals (~10–80 mmol/L). "Salty sweaters" — visible salt streaks on dark clothing — have a real basis and a real need for higher exercise sodium intake.
Pregnancy: sodium needs rise modestly; magnesium needs rise (RDA 350–400 mg).
Children: scaled-down RDAs; the entry targets adults.
Athletes vs. sedentary: the entire exercise-fueling case applies to athletes and not to sedentary readers.
Knowledge gaps
What hasn't been settled: the optimal sodium intake for normotensives at moderate baseline — PURE J-curve vs. Aburto monotonic remains contested. Whether the very-low (<1.5 g/day) sodium tier provides incremental benefit beyond the moderate (~2.3 g/day) tier outside of severe hypertension. Whether magnesium supplementation in non-deficient adults provides any measurable benefit — most positive trials enrolled deficient subjects. Long-term safety of high-dose magnesium threonate (the newer brain-bioavailable form marketed for cognition). Whether SSaSS-style salt substitution generalizes outside the high-sodium, high-CV-risk rural Chinese population studied. Whether endurance "electrolyte" products meaningfully outperform plain sodium + water at matched dose. Causal role of magnesium status in migraine prophylaxis (mechanism plausible, trials mixed). What evidence would change the author's call: a large RCT testing moderate (2.3 g) vs. very low (1.5 g) sodium on hard CV endpoints in moderate-risk adults; a definitive magnesium-for-sleep trial in middle-aged non-insomniacs; a salt-substitute trial in Western dietary patterns.
Scope. The brief named six consequences (hydration, BP, cramps, exercise performance, sleep, cardiac rhythm). All six are surfaced. Cardiac rhythm is addressed through the mechanism section (Na/K-ATPase, potassium and magnesium as repolarization determinants) and the contraindications callouts (digoxin/RAAS-inhibitor interactions) rather than getting a standalone section — the felt-experience hook for a healthy reader is BP and sleep, not arrhythmia, so the editorial weight tilts there. Hydration is addressed in protocol, audience (endurance subgroup), and misconceptions (exercise-associated hyponatremia) rather than dwelling on the general "drink water" case, which belongs in a separate hydration entry.
Calcium excluded. The fourth canonical electrolyte. Its dominant story is bone density and warrants a separate entry under skin/lookmaxxing or supplements. Flagged in out-of-scope.
Cramps framed against the gym-floor consensus. Schwellnus 2009 and the 2020 Cochrane Garrison review point against the depletion theory. The article calls this out honestly in misconceptions. Some readers will push back; the evidence is on the article's side.
Focus and mood scored 0. Initially considered 1 each — mechanism for magnesium on mood (GABA-A, NMDA) is plausible, and severe dehydration impairs cognition — but the article doesn't dedicate a paragraph to either, and the evidence in non-deficient adults is too thin to justify a non-zero score without article coverage. Honest 0 is cleaner than padded 1. If a future trial lands a magnesium-mood result in non-deficient adults, revisit.
Rating difficulty: evidence at 4 not 5. SSaSS is hard-endpoint RCT-grade; DASH-Sodium and Aburto meta-analyses are strong. But PURE's J-curve and SODIUM-HF's null result are genuine unresolved heterogeneities. Controversy at 3 reflects the same.
Rating difficulty: longevity at 4. SSaSS's 12% all-cause mortality reduction over 5 years carries the score. Generalizability to lower-risk Western populations is debated — flagged in knowledge gaps. 4 not 5 because the load-bearing trial is in a high-CV-risk cohort and replication outside it isn't yet in.
Rating difficulty: beauty_cumulative at 1. Indirect, via BP/vascular health on long-term skin microcirculation and arterial-wall aging. Real but small; could defensibly be 0. Kept at 1 because the stakes/payoff prose explicitly traces the vascular-trajectory story.
Contraindications. Used kidney-disease, cardiac-condition (covering HF and arrhythmia interactions with K and Mg), blood-thinners (warfarin patients monitoring dietary K). Did not use uncontrolled-hypertension — the entry is largely for hypertensives.
Future-link candidates (flagged in out-of-scope so a future editor can wire them when the entries exist):
- A standalone DASH entry
- A calcium-and-bone entry
- A home BP monitoring entry
- An endurance-fueling entry for athletes
- A general water-intake entry
Separate-entry candidate. "Salt sensitivity" is a real biological axis (~50% of hypertensives, ~25% of normotensives) that surfaces here but doesn't get its own treatment. Could merit a stub entry under screening.
Audience scoping. Did not set audience.gender or audience.ages on meta — the entry applies to all adults, with subgroup emphasis surfaced in the article's audience section instead. Setting an age band would suppress the entry for younger readers, who should still see the exercise-fueling case.
Hard call on keto/paleo voice. The "keto flu = sodium deficit" point is accurate; the broader "salt is harmless at any intake" influencer narrative is not. Included the keto subgroup in audience to keep the entry's coverage honest without endorsing the overreach.
Electrolytes
Salt substitutes ~$5-15/year, magnesium glycinate ~$10-25/month, dietary potassium from food is roughly cost-neutral. Annual cost typically well under $300.
SSaSS hard-endpoint RCT: 14% stroke reduction, 13% major CV events, 12% all-cause mortality over 5 years from a single salt substitution (Neal 2021). Aburto BMJ meta-analyses confirm independent sodium and potassium effects on BP and stroke. One of the higher-impact dietary levers in the catalogue when applied to the right population.
Shifting away from processed/restaurant food is a real lifestyle change — the dominant sodium source is hidden in packaged food (Cogswell 2016). Adding potassium-rich foods and an evening magnesium dose is modest but sustained daily effort.
SSaSS is hard-endpoint RCT-grade with mortality reduction (Neal 2021). DASH-Sodium, Aburto meta-analyses, Cochrane reviews, NASEM DRIs, and WHO guidelines align on the core message. Magnesium evidence is meta-analysis-grade (Zhang 2016 BP; Mah 2021 sleep). Not 5 because SODIUM-HF nulled aggressive restriction in HF and PURE's J-curve remains unresolved.
Clear functional improvements within weeks: ~5-7/3 mmHg BP drop in hypertensives on DASH-Sodium within 2 weeks (Sacks 2001); sleep-onset improvement in deficient older adults on magnesium (Abbasi 2012, Mah 2021); reduced cramping when sodium replaced during prolonged sweating (McDermott 2017).
Adequate hydration plus sodium replacement during prolonged exercise produces a real performance and recovery lift (McDermott 2017). Outside athletes, day-to-day energy effects are smaller and largely downstream of better sleep and BP control.
Magnesium 200-500 mg/day improves sleep-onset latency by ~17 minutes in older adults with insomnia (Abbasi 2012); 2021 systematic review confirms modest effect in this subgroup (Mah 2021). Works in deficient/older populations primarily.
Adequate BP control and vascular health contribute marginally to long-term appearance through the artery-wall and skin-microcirculation axes; no direct cosmetic mechanism. Indirect, slow, real.