Lab Library

Cardiovascular

Cardiovascular is a lab panel for heart and vascular risk beyond standard cholesterol -- ApoB, Lp(a), inflammation, and the metabolic drivers behind them. It covers 16 markers. This page explains what each one measures, what it does NOT measure, and which nutrients are known to move it.

The markers, and what they actually tell you

Each marker below carries its technical name, what it measures in plain language, the limits of what it can tell you, and the nutrients known to move it. Where a framework and mainstream practice disagree, both views are shown rather than one being hidden.

ApoB (Apolipoprotein B)

A direct headcount of the atherogenic particles in your blood. Every particle that can lodge in an artery wall -- LDL, VLDL, IDL, and Lp(a) -- carries exactly one ApoB molecule on its surface, so a single ApoB test counts all of them at once. Where LDL-C estimates the cholesterol cargo, ApoB counts the containers, which is why it can flag a high particle burden hiding under a "normal" cholesterol.

Typical range: Conventional: roughly 40-125 mg/dL (assay-dependent, and interpreted against risk category rather than a single universal cutoff). Preventive-cardiology practitioners often cite lower targets (commonly quoted around <80 mg/dL for higher-risk individuals, or lower still) -- these are risk-stratified clinical opinions to discuss with a clinician, not a universal lab "optimal" line, and ApoB is meant to be read alongside the rest of the picture, not in isolation.

Food first: The food-first levers act on the whole particle picture, not ApoB alone: soluble fiber (oats, legumes, vegetables, fruit), replacing refined carbohydrate and excess saturated fat with unsaturated fats and whole plants, and addressing insulin resistance. There is no evidence-based standalone supplement that reliably lowers ApoB.

When to see someone: ApoB is the better particle-count marker, but a single value is interpreted by risk category and alongside the rest of the panel, not as a stand-alone verdict. Do not treat it as something a smoothie or supplement fixes -- the honest framing is measure it, know your particle burden, and work the whole-diet and metabolic levers with a clinician.

Read the studies on ApoB (Apolipoprotein B)

Lp(a) (Lipoprotein(a))

An LDL-like particle with an extra protein strand wrapped around it, whose level is set overwhelmingly by your genes and is largely independent of diet, exercise, and the usual lifestyle levers. Because it is stable across life, it is a measure-once marker: a single test tells you whether you carry an inherited layer of cardiovascular risk that no standard lipid panel would ever reveal.

Typical range: Conventional: commonly reported as desirable below ~30 mg/dL (or <75 nmol/L), with rising risk above that -- but units and assays vary substantially (mg/dL vs nmol/L are not interchangeable), so always read the specific assay's reference range. There is no "functional optimal" you move toward with habits, because the level is largely inherited.

Food first: There is no reliable diet or supplement lever for Lp(a) itself. The evidence-based response to a high result is not a supplement chase -- it is to take every CHANGEABLE lever more seriously (particle number, blood pressure, inflammation, metabolic health) and to treat a high Lp(a) as a clinician conversation about your whole risk picture.

When to see someone: Measure once to know your inherited baseline; do not re-test repeatedly expecting movement, and do not let anyone sell you a product to "lower your Lp(a)" through diet. A high value is a signal to manage overall risk with a clinician, not a number to self-manage.

Read the studies on Lp(a) (Lipoprotein(a))

LDL-C (LDL Cholesterol)

An estimate of how much cholesterol CARGO is riding inside all your LDL particles combined -- a measure of contents, not a count of particles. Because the artery wall responds to how many particles bump into it rather than how full each one is, LDL-C can read "normal" while the particle number (ApoB) is high, especially in people carrying many small, dense particles.

Typical range: Conventional: often reported as "optimal <100 mg/dL," "near optimal 100-129," rising from there, though treatment targets are individualized by overall risk. Note LDL-C is frequently CALCULATED (Friedewald equation) rather than directly measured, and the calculation becomes unreliable when triglycerides are high -- a directly measured or ApoB value is more trustworthy in that setting.

Nutrients that move it

  • Soluble fibersolid -- dietary fiber lowers cardiovascular risk in its own right and improves the lipid picture; pooling prospective studies, each additional 7 grams of daily fiber was associated with roughly a 9% lower risk of cardiovascular disease, partly through its effect on lipids and blood sugar. (cite: Dietary fibre intake and risk of cardiovascular disease: systematic review and meta-analysis, 2013, PMID 24355537)

Food first: Soluble fiber (oats, barley, legumes, apples, citrus), replacing excess saturated fat with unsaturated fats (olive oil, nuts, fatty fish), and a plant-forward plate. These act on the whole lipid and particle picture.

When to see someone: A "normal" LDL-C is a start, not an answer -- pair it with ApoB when you want to know the real particle burden. When triglycerides are high, a calculated LDL-C can be misleading; ask about a directly measured LDL or an ApoB instead.

Read the studies on LDL-C (LDL Cholesterol)

HDL-C (HDL Cholesterol)

Cholesterol carried inside HDL particles, involved in returning cholesterol from tissues back toward the liver. Higher HDL-C is generally read as favourable, but like LDL-C it is a cargo measure, not the whole story -- and very high HDL from some causes does not carry the protective meaning people assume.

Typical range: Conventional: roughly >40 mg/dL (men) and >50 mg/dL (women) as the low-cutoff for the metabolic-syndrome definition; higher is generally read as favourable within reason. There is no validated "push it as high as possible" target -- HDL is most useful read as part of the pattern (especially the triglyceride:HDL ratio) rather than maximized in isolation.

Nutrients that move it

  • Omega-3 fats (EPA/DHA)emerging -- omega-3 fats become part of the cell membranes that shape endothelial signalling and are foundational membrane and triglyceride nutrients, though their effect is on the triglyceride and membrane side rather than a reliable HDL-raising lever, and general supplementation in broad populations has not consistently reduced cardiovascular events. (cite: Omega-3 supplementation and cardiovascular disease: meta-analysis, 2012, PMID 22968891)

Food first: Physical activity is the most reliable everyday lever on HDL; a plant-forward, fiber-rich diet, unsaturated fats, and addressing insulin resistance support it indirectly. No standalone supplement is an evidence-based HDL raiser.

When to see someone: Do not chase HDL as a number to maximize -- context matters, and a "good HDL" does not cancel out a high ApoB or a high triglyceride:HDL ratio. Read it as one piece of the pattern.

Read the studies on HDL-C (HDL Cholesterol)

Triglycerides

The main fat circulating in the blood, measured fasting for the most interpretable result. High triglycerides commonly travel with insulin resistance and small, dense LDL particles, which is why triglycerides are as much a metabolic signal as a lipid one.

Typical range: Conventional: normal <150 mg/dL, borderline 150-199, high 200-499. Fasting matters -- a non-fasting sample runs higher. Some preventive practitioners prefer well under 100 mg/dL as a "metabolically clean" opinion target, which is opinion, not a diagnostic cutoff.

Nutrients that move it

  • Omega-3 fats (EPA/DHA)solid -- high-dose EPA in people with elevated triglycerides was associated with fewer cardiovascular events in a large trial, and omega-3 fats are a well-established triglyceride-lowering nutrient; the honest caveat is that general omega-3 supplementation in broad, unselected populations has NOT consistently reduced cardiovascular events in meta-analysis. So omega-3s are foundational triglyceride and membrane nutrients with a real signal in the right context, not a magic pill for everyone. (cite: Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia, REDUCE-IT, 2019, PMID 30415628; Omega-3 supplementation and cardiovascular disease: meta-analysis, 2012, PMID 22968891)
  • Soluble fibersolid -- fiber improves the lipid and blood-sugar picture triglycerides sit within; each additional 7 g/day of fiber was associated with roughly 9% lower cardiovascular disease risk. (cite: Dietary fibre intake and risk of cardiovascular disease, 2013, PMID 24355537)

Food first: The biggest food levers on triglycerides are the metabolic ones: less added sugar and refined carbohydrate, less alcohol, more fiber and movement, and omega-3 fats from fatty fish (food-level intake is the sensible base). This overlaps heavily with the insulin-resistance levers in the Metabolic Drivers section.

When to see someone: Very high triglycerides (typically >500 mg/dL) are a distinct medical situation (pancreatitis risk) that needs a clinician, not a diet-only approach. A non-fasting draw inflates the number -- confirm fasting before acting on a borderline value.

Read the studies on Triglycerides

Total Cholesterol

The sum of cholesterol carried across all lipoprotein particles (roughly HDL-C + LDL-C + a fraction of triglyceride-rich particles). It is the crudest number on the panel because it blends "favourable" and "atherogenic" cholesterol into one figure, so a total cholesterol read in isolation can be actively misleading.

Typical range: Conventional: desirable <200 mg/dL, borderline 200-239, high ≥240. But a total cholesterol is nearly meaningless without its breakdown -- a high total driven by high HDL is not the same story as one driven by high LDL/particle number.

Food first: Same as LDL-C and triglycerides -- a fiber-rich, plant-forward plate with unsaturated fats. But interpret the change through the breakdown (LDL-C, HDL-C, ApoB), not the total alone.

When to see someone: Never act on total cholesterol without its components and, ideally, ApoB -- a "high total cholesterol" can be benign (high HDL) or a real signal (high particle number), and the total cannot tell you which.

Read the studies on Total Cholesterol

Non-HDL Cholesterol

Total cholesterol minus HDL-C -- a derived number (no extra blood draw needed) that captures the cholesterol carried by ALL the atherogenic particles (LDL, VLDL, IDL, remnants, Lp(a)) in one figure. It is a cheap, always-available proxy for ApoB, and unlike calculated LDL-C it stays reliable even when triglycerides are high.

Typical range: Conventional: often cited as roughly 30 mg/dL above the corresponding LDL-C goal (e.g., a non-HDL "optimal" around <130 mg/dL when LDL goal is <100), individualized by overall risk. Because it is derived from numbers already on the report, it is free information.

Food first: Same whole-diet levers as ApoB and LDL-C -- soluble fiber, unsaturated fats replacing saturated/refined, and metabolic health.

When to see someone: Non-HDL is a strong, cheap proxy for particle burden and more robust than calculated LDL-C when triglycerides are high -- but where available, a direct ApoB is still the better particle-count marker. Read it as part of the pattern, not alone.

Read the studies on Non-HDL Cholesterol

Triglyceride:HDL Ratio

Triglycerides divided by HDL cholesterol -- a derived number (free from values already on a standard panel) that reads as a proxy for insulin resistance, the state in which cells respond sluggishly to insulin. A high ratio is the fingerprint of the small, dense LDL particles and the metabolic pattern that go with insulin resistance, making it a bridge between the lipid panel and the blood-sugar story.

Typical range: No official clinical reference range exists; it is a proxy, not a diagnostic test. Commonly cited "favourable" thresholds hover around <2 (using mg/dL for both, in US units), with higher ratios suggesting more insulin resistance -- but the exact cutoff varies by population and units (mg/dL vs mmol/L differ), so treat any specific number as a heuristic, not a validated cutoff. In a study of non-diabetic overweight adults, fasting triglycerides and this ratio identified insulin-resistant individuals about as well as more elaborate testing.

Nutrients that move it

  • Magnesiumemerging -- magnesium is mechanistically involved in insulin signalling and glucose handling, so magnesium sufficiency plausibly supports the insulin-sensitivity side this ratio tracks; the strongest food-first case is correcting a real deficiency rather than empiric high-dose supplementation (dedicated trials isolating magnesium's effect on this specific ratio are not the basis of this entry).

Food first: The blood-sugar levers move this ratio: fiber, movement, less ultra-processed food and added sugar, less alcohol, and weight management if relevant -- not stacking supplements on top of an unaddressed metabolic pattern.

When to see someone: This is a proxy, not a diagnosis of insulin resistance -- confirm the metabolic picture with fasting insulin, glucose, and HbA1c. Unit systems change the "good" number, so calculate it consistently and compare to your own trend.

Read the studies on Triglyceride:HDL Ratio

hs-CRP (High-Sensitivity C-Reactive Protein)

Cellular Six: Maintain

This marker sits in Maintain -- the cleanup and defence system, and whether it is switched on.

A marker of low-grade, whole-body inflammation. It matters on a cardiovascular panel because atherosclerosis is not only a plumbing problem of cholesterol build-up -- it is also an inflammatory process in the vessel wall. In nearly 28,000 women followed for years, baseline hs-CRP was a stronger predictor of cardiovascular events than LDL cholesterol.

Typical range: Conventional cardiovascular risk stratification: low risk <1.0 mg/L, average 1.0-3.0 mg/L, higher-risk >3.0 mg/L. A value above ~10 mg/L usually reflects acute inflammation (infection, injury) rather than baseline vascular risk and should prompt a repeat once well.

Nutrients that move it

  • Omega-3 fats (EPA/DHA)emerging -- omega-3 fats are part of the membrane and inflammatory-signalling machinery and are foundational anti-inflammatory nutrients, though, as with cardiovascular events, broad omega-3 supplementation has not shown consistent benefit in unselected populations, so the honest framing is food-first fatty fish rather than an assumed hs-CRP-lowering pill. (cite: Omega-3 supplementation and cardiovascular disease: meta-analysis, 2012, PMID 22968891)
  • Fiber and a plant-forward dietsolid -- fiber-rich, whole-plant diets are associated with lower cardiovascular risk and a lower inflammatory burden; the strongest peer-reviewed anchor here is fiber's cardiovascular association rather than an hs-CRP endpoint specifically. (cite: Dietary fibre intake and risk of cardiovascular disease, 2013, PMID 24355537)

Food first: The whole anti-inflammatory pattern moves hs-CRP more than any single nutrient: fiber and polyphenol-rich plants (berries, cocoa, vegetables), omega-3 fats from fatty fish, less ultra-processed food and added sugar, adequate sleep, and movement.

When to see someone: hs-CRP is non-specific -- it rises with any inflammation, from a cold to a sore knee -- so it is read as a trend and in context, never as a stand-alone verdict. Don't interpret a single elevated value as cardiovascular risk without ruling out an acute cause and repeating it when well.

Read the studies on hs-CRP (High-Sensitivity C-Reactive Protein)

Homocysteine

Cellular Six: Build

This marker sits in Build -- the raw material for making and renewing structure.

An amino acid the body is meant to clear, and clearing it depends on methylation -- a cellular housekeeping process run by the B vitamins B6, folate (B9), and B12. When those cofactors run short, homocysteine climbs, which is why an elevated level is read partly as a methylation- and B-vitamin-status signal.

Typical range: Conventional: roughly 5-15 micromol/L, with many labs flagging above ~15. Some practitioners prefer a lower "optimal" (often quoted under ~7-10 micromol/L) as a methylation-status opinion target -- this is opinion, not a validated diagnostic cutoff, and it is read as a B-vitamin/methylation signal rather than a number to chase for its own sake.

Nutrients that move it

  • B6, folate (B9), and B12solid -- decades of randomized trials confirm B6, B9, and B12 lower blood homocysteine through their roles as methylation cofactors; but here is the honesty the marker demands: lowering homocysteine with B vitamins has NOT reliably reduced cardiovascular events in large trials -- a five-year vascular-disease trial lowered homocysteine with folic acid, B6, and B12 and still saw no drop in events. So these B vitamins reliably move the number, and the number is a useful read on methylation and B-vitamin status, but it is not a lever guaranteed to change outcomes when you push on it. (cite: Effects of B Vitamins on Homocysteine Lowering and Thrombotic Risk Reduction -- A Review of Randomized Controlled Trials, 2024, PMC11990291; Homocysteine Lowering with Folic Acid and B Vitamins in Vascular Disease, HOPE-2, 2006, NEJM)

Food first: Folate from leafy greens, legumes, and fortified grains; B12 from animal foods (or a supplement for those who eat little animal food, older adults, or people on acid-suppressing medication); B6 from a varied diet. Correcting a genuine B-vitamin shortfall is the defensible reason to supplement, not chasing the number for its own sake.

When to see someone: A high homocysteine is informative about methylation and B-vitamin status, but the trial evidence says lowering it does not reliably reduce cardiovascular events -- so don't treat the number as a target to hit at any cost. High-dose folic acid can mask a B12 deficiency, so B12 status should be checked before loading folate.

Read the studies on Homocysteine

Fasting Insulin

The amount of insulin circulating after an overnight fast. It is one of the earliest signals of insulin resistance -- insulin often climbs to keep glucose normal for years before fasting glucose itself starts to rise, so a fasting insulin (and the derived HOMA-IR, from fasting glucose and insulin) can reveal a metabolic pattern that a normal glucose hides.

Typical range: Conventional: roughly 2-25 microIU/mL (wide, assay-dependent). Many metabolic-health practitioners prefer a much lower "optimal" (often quoted single digits, e.g., under ~5-8 microIU/mL) as an insulin-sensitivity opinion target -- this is opinion, not a diagnostic cutoff. HOMA-IR is calculated from fasting glucose and insulin as a resistance estimate.

Nutrients that move it

  • Magnesiumemerging -- magnesium is a cofactor in insulin signalling and glucose handling, so magnesium sufficiency plausibly supports insulin sensitivity; the food-first case is correcting a real shortfall rather than empiric high-dose supplementation. Insulin resistance itself degrades the endothelium by shifting it away from nitric oxide, which is why the metabolic markers belong on a cardiovascular panel. (cite: Role of Insulin Resistance in Endothelial Dysfunction, 2013, PMC3594115)

Food first: Fiber, whole plants, protein with meals, less added sugar and refined carbohydrate, movement (especially after meals), and adequate sleep. These are the same levers that move the triglyceride:HDL ratio.

When to see someone: Fasting insulin assays vary widely, so compare to your own trend and the same lab. It must be drawn fasting to be interpretable, and it is read alongside glucose and HbA1c, not alone.

Read the studies on Fasting Insulin

Fasting Glucose

Blood sugar after an overnight fast -- a snapshot of glucose control that changes later than fasting insulin. A normal fasting glucose does not rule out early insulin resistance (insulin may already be working overtime to keep it there), which is why it is read with fasting insulin and HbA1c.

Typical range: Conventional: normal <100 mg/dL, "prediabetes" range 100-125 mg/dL, with ≥126 mg/dL on repeat testing meeting the diabetes threshold (a clinical diagnosis for a physician, not a self-diagnosis). Some practitioners prefer the lower end of normal as an opinion target.

Nutrients that move it

  • Magnesiumemerging -- magnesium's mechanistic role in glucose handling makes sufficiency supportive of normal glucose regulation; correcting a genuine deficiency is the food-first lever, not empiric megadosing.

Food first: Fiber, whole plants, protein and fat with carbohydrate to blunt spikes, less added sugar and refined carbohydrate, movement after meals, and adequate sleep.

When to see someone: Diagnosing diabetes or prediabetes from a fasting glucose is a clinical judgment (usually confirmed with a repeat test or HbA1c) for a physician -- a single value is a signal, not a diagnosis. Must be drawn fasting.

Read the studies on Fasting Glucose

HbA1c (Hemoglobin A1c)

The share of hemoglobin that has glucose stuck to it, which reflects average blood sugar over roughly the prior 2-3 months (the lifespan of red blood cells). It smooths out the day-to-day swings a single glucose reading shows, making it the standard longer-term glucose-control marker.

Typical range: Conventional: normal <5.7%, "prediabetes" range 5.7-6.4%, ≥6.5% meeting the diabetes threshold (a clinical diagnosis). Some metabolic-health practitioners prefer the lower end of normal as an opinion target.

Nutrients that move it

  • Magnesiumemerging -- magnesium sufficiency supports the glucose-handling machinery HbA1c reflects; correct a genuine shortfall rather than megadosing.

Food first: Sustained dietary change (fiber, whole plants, fewer refined carbohydrates and added sugars) and regular movement, since HbA1c reflects months of average glucose -- there is no quick nutrient fix that moves it in days.

When to see someone: HbA1c can read falsely low with conditions that shorten red-cell lifespan (some anemias, recent blood loss) and falsely high in others -- interpret an unexpected value with a clinician and, if needed, cross-check against fasting glucose. It reflects an average, so it can hide large glucose swings around a normal-looking mean.

Read the studies on HbA1c (Hemoglobin A1c)

Potassium

Cellular Six: Transform

This marker sits in Transform -- the cofactors enzymes need to turn fuel into usable energy.

The main intracellular electrolyte, tightly regulated in the blood. On a cardiovascular panel its everyday relevance is the sodium-to-potassium balance: potassium supports the endothelium and healthy vascular tone, and raising potassium intake measurably lowers blood pressure. The modern diet inverts the ancestral ratio -- heavy on sodium from processed food, light on potassium from plants.

Typical range: Conventional serum: roughly 3.5-5.0 mmol/L (tightly regulated -- serum potassium is a safety marker, not a nutrition-intake gauge, since the body defends blood levels even when dietary intake is low). The meaningful "target" is dietary intake (adequate intake commonly cited around 2,600-3,400 mg/day for adults), not pushing the serum number.

Nutrients that move it

  • Potassium (dietary)solid -- pooled trials found increased potassium intake reduced systolic blood pressure, with the largest effect in people with hypertension, and potassium supports the endothelium and healthy vascular tone. The lever is dietary potassium from vegetables and fruit, and it acts on the vessel lining, not just a blood-pressure cuff reading. (cite: Effect of increased potassium intake on cardiovascular risk factors and disease: systematic review and meta-analyses, 2013, PMID 23558164)

Food first: Potassium from food -- vegetables, fruit, legumes, and tubers -- rather than a pill. Shifting the Na:K balance toward potassium (more plants) and away from sodium (less packaged food) is the most everyday lever on the endothelium there is.

When to see someone: This is about DIETARY potassium in people with healthy kidneys. Potassium interpretation and intake targets REVERSE in diagnosed kidney disease, and potassium supplements or salt substitutes can be dangerous for people with impaired kidney function or on certain blood-pressure medications (ACE inhibitors, ARBs, potassium-sparing diuretics) -- those individuals need a clinician's guidance, not a general "eat more potassium" message. A serum potassium outside range (high or low) is a medical finding to address with a clinician, not with diet alone.

Read the studies on Potassium

Sodium (and the Na:K Ratio)

The main extracellular electrolyte and, in the diet, the other half of the sodium-to-potassium balance. Serum sodium is tightly regulated and mostly reflects fluid balance rather than dietary intake, so the cardiovascular story here is about DIETARY sodium (largely from processed and restaurant food) and the Na:K ratio, not the blood value.

Typical range: Conventional serum: roughly 135-145 mmol/L (a fluid-balance and safety marker, not a diet gauge). The meaningful dietary framing is the Na:K ratio: the modern diet is heavy on sodium and light on potassium, and shifting it back toward potassium supports healthy blood pressure and the vessel lining. Common dietary sodium guidance sits around or below ~2,300 mg/day, lower for many adults.

Nutrients that move it

  • Na:K balancesolid -- the best-evidenced half of the ratio is that raising potassium intake lowers systolic blood pressure; potassium softens and supports the endothelium while a sodium-heavy, potassium-poor diet works against it. The practical lever is the ratio, not sodium in isolation. (cite: Effect of increased potassium intake on cardiovascular risk factors and disease, 2013, PMID 23558164)

Food first: Cut sodium from ultra-processed and restaurant food (the dominant dietary source, far more than the salt shaker) and raise potassium from vegetables and fruit -- both moves shift the Na:K ratio in the same favourable direction at once.

When to see someone: A serum sodium outside range is a fluid-balance/medical finding, not a diet target -- do not try to "correct" a blood sodium with diet. Salt substitutes replace sodium with potassium, which is helpful for many but risky for people with kidney disease or on potassium-affecting medications (see Potassium caveats). Very aggressive sodium restriction is not automatically better for everyone; the balanced, plant-forward Na:K shift is the general-population lever.

Read the studies on Sodium (and the Na:K Ratio)

Magnesium

Cellular Six: Transform

This marker sits in Transform -- the cofactors enzymes need to turn fuel into usable energy.

A mineral involved in over 300 enzymatic reactions, including insulin signalling, glucose handling, and vascular tone, so it works alongside potassium in blood-pressure and endothelial regulation. Serum magnesium reflects less than 1% of total body magnesium and is an imperfect status marker; RBC magnesium is often preferred by practitioners as more informative (though it is not a routine panel test).

Typical range: Conventional serum: roughly 1.7-2.2 mg/dL (a poor reflection of total-body stores). Some practitioners prefer RBC magnesium and target its upper range as an opinion; RDA is roughly 400-420 mg/day (men) and 310-320 mg/day (women).

Nutrients that move it

  • Magnesiumemerging -- magnesium works alongside potassium in blood-pressure and vascular regulation and is a cofactor in the insulin signalling that shapes the metabolic side of cardiovascular health; correcting a genuine deficiency is the food-first lever, and the insulin-resistance-to-endothelium link is why it matters vascularly. (cite: Role of Insulin Resistance in Endothelial Dysfunction, 2013, PMC3594115)

Food first: Leafy greens, nuts, seeds, legumes, whole grains, and dark chocolate; magnesium glycinate or citrate are generally better absorbed and tolerated than oxide when a supplement is warranted for a confirmed shortfall.

When to see someone: High-dose magnesium supplements cause diarrhea before serious harm in people with normal kidney function, but anyone with reduced kidney function (low eGFR) needs medical guidance before supplementing, since magnesium is renally cleared. Serum magnesium can look "normal" while total-body stores are low, so a normal serum value does not rule out a shortfall.

Read the studies on Magnesium

Food first, then fill the gaps

Every marker above lists the nutrients known to move it. Build a day around them first -- the Diet Builder shows which food groups you are consistently missing, inside the way you already eat.