Does an alkaline diet change your blood pH?

By about 0.014 units. That figure comes from a four-period double crossover in which the same eight people ate deliberately acid-forming and base-forming diets matched for calcium, phosphate, sodium, protein and calories. Urine pH in the same study differed by 1.02 units. Urine pH is what the diet changes; blood pH is what the cancer claim needs, and the difference between those two numbers is the whole argument.

Does it protect your bones?

The first half of the mechanism replicates and the second does not. Raising the diet acid load reliably raises urine calcium, at about 66 mg a day per 47 mEq change. But across the five studies meeting the strictest methodological standard, there was no relationship between change in net acid excretion and change in whole body calcium balance, at p = 0.38 with 94 percent power. The calcium in the urine does not appear to be a net loss from the body.

What about the trial where potassium citrate increased bone density?

It is real and it belongs in any honest account. 161 postmenopausal women with low bone mass were randomised to potassium citrate or potassium chloride for twelve months, and the citrate arm gained 1.87 percent at the lumbar spine, 1.39 at the femoral neck and 1.98 at total hip, all p < 0.001. Both arms got potassium, so the comparison isolates the alkali. The complication is that a separate two-year trial tested potassium citrate at two doses alongside 300 g a day of extra fruit and vegetables and found no difference in bone density or turnover from placebo.

There is no trial. A systematic review screened 8,278 citations and reviewed 252 abstracts. One study met the inclusion criteria, no randomised trials were located, and no studies of any kind were found on dietary acid load or alkaline water as a cancer treatment. The one included cohort found no association between diet acid load and bladder cancer. That is not a demonstration that the claim is false; it is a demonstration that after an exhaustive search there is essentially nothing to evaluate.

Elimination4 human studies cited

Alkaline diet

Q: Can it treat cancer?

There is no trial. A systematic review screened 8,278 citations and reviewed 252 abstracts. One study met the inclusion criteria, no randomised trials were located, and no studies of any kind were found on dietary acid load or alkaline water as a cancer treatment. The one included cohort found no association between diet acid load and bladder cancer. That is not a demonstration that the claim is false; it is a demonstration that after an exhaustive search there is essentially nothing to evaluate.

Q: Should I cut protein to lower my acid load?

The evidence that would justify that is the evidence that failed. The balance meta-analyses found no sign that dietary acid load causes skeletal mineral loss, and the causal assessment found randomised studies providing no evidence for an adverse role of phosphate, milk or grain foods in osteoporosis. The clinical review of this specific question states the opposite concern: in the elderly, inadequate protein intake is a greater problem for bone health than protein excess.

Written by Aaron CuhaReviewed Sep 2026

Also called: Acid-alkaline diet, Acid-ash diet, pH diet, Alkaline water

Half the mechanism replicates and half does not. Raise the diet acid load and urine calcium rises, reliably. Whole body calcium balance does not change, at 94 percent power, and no study has found a biological mechanism operating at physiological pH.

What it actually is

Reduce foods classified as acid-forming, principally meat, fish, hard cheese and grains, and raise foods classified as alkaline-forming, principally fruit and vegetables, often alongside alkaline water or alkaline salts. The classification itself is real and measurable. The potential renal acid load of a food is calculated from its nutrient composition and runs from an average maximum around plus 23.6 mEq per 100 g for some hard cheeses, through 0 for fats and oils, to about minus 3 for fruits, fruit juices and vegetables, and the resulting predicted net acid excretion correlates with urine pH at r = 0.83. So the food categories are not invented. What is disputed is what changing them does.

The mechanism its advocates propose

Two claims of very different strength, and they should be separated before either is tested. The acid-ash hypothesis holds that the modern diet delivers a chronic acid load, that the body buffers it by mobilising alkaline mineral salts from the skeleton, that the liberated calcium appears in urine, and that over decades this causes osteoporosis. That is a specific physiological hypothesis with clear predictions and it has been tested repeatedly in metabolic balance studies. The second and stronger claim is that an alkaline diet raises systemic pH and thereby prevents or treats cancer. That one requires arterial pH to move, and arterial pH is held between roughly 7.35 and 7.45 by respiratory and renal compensation more or less regardless of what is eaten, which is why urine pH and blood pH are different measurements and why almost every trial here reports the first.

What happened when calories were controlled

Strongest design available: controlled

Energy intake was matched between groups, so a difference in the result is attributable to the composition of the diet rather than to how much was eaten.

Calories are not the disputed variable here; acid load is, and it has been controlled repeatedly in metabolic balance studies with calcium, phosphate, sodium, protein and energy matched between arms. The first half of the mechanism holds up: raise the acid load and urine calcium rises. The second half fails: whole body calcium balance does not change.

Fenton 2008 pooled 25 of 105 studies to quantify the effect. Net acid excretion on control diets averaged 47 mEq/day, and a change of 47 mEq in net acid excretion was associated with 1.6 mmol/day, or 66 mg/day, of urine calcium in acidic urine, with a significant linear relation in both acidic and alkaline urine (p < 0.001). The authors' own qualifier is the load-bearing sentence: this is not evidence that the source of the excreted calcium is bone, or that the calciuria contributes to osteoporosis.
Fenton 2009 in the Journal of Bone and Mineral Research tested exactly that, restricted to studies meeting the Institute of Medicine's methodological standards for calcium balance. Five of 16 studies qualified. There was a significant linear relationship between rising net acid excretion and urine calcium (p < 0.0001), no relationship between a change in net acid excretion and a change in calcium balance (p = 0.38 at 94 percent power), and no relationship with the bone resorption marker N-telopeptides (p = 0.95). A null at 94 percent power is a much stronger statement than a null usually is.
A companion meta-analysis attacked the hypothesis through phosphate, which the acid-ash argument names as a major contributor from dairy, meat and grains. Twelve studies, 30 intervention arms, 269 subjects. Higher phosphate intakes decreased urine calcium and increased calcium retention, at both high and low calcium intakes and regardless of the protonation of the phosphate supplement. Every analysis ran contrary to the hypothesis.
Buclin 2001 is the trial that supports the mechanism and it belongs here in full. Eight healthy volunteers, four periods in a double crossover, acid-forming against base-forming diets matched for calcium, phosphate, sodium, protein and calories. Blood pH differed by an average of 0.014 (p = 0.002) and urine pH by 1.02 (p < 0.0001), which is the cleanest available illustration of how the two measurements differ in scale. Urine calcium was 74 percent higher on the acid-forming diet (p < 0.0001) and C-telopeptide excretion 19 percent higher (p = 0.01), which the authors read as a skeletal origin for the excess calcium. Eight people and four days of dietary preparation.
Fenton 2011 applied Hill's criteria for causality across 55 of 238 studies, including 22 randomised interventions and 11 prospective cohorts. Urine calcium excretion rates were consistent with the hypothesis and calcium balance studies did not demonstrate whole body calcium loss with higher net acid excretion. No intervention study provided direct evidence of osteoporosis progression, meaning fragility fractures or bone strength measured by biopsy. The supporting cohorts were not controlled for weight loss during follow-up, family history of osteoporosis, baseline bone mineral density or oestrogen status. No study revealed a biologic mechanism functioning at physiological pH, and randomised studies did not provide evidence for an adverse role of phosphate, milk or grain foods in osteoporosis.

What happened when people just ate it

The fair real-world test is the randomised alkali trials, because alkaline salts are exactly what the diet is trying to deliver through food, and here the literature genuinely disagrees with itself. Macdonald 2008 randomised 276 postmenopausal women aged 55 to 65 to four arms for two years: potassium citrate at 55.5 mEq/day, potassium citrate at 18.5 mEq/day, placebo, or 300 g/day of additional fruit and vegetables, which supplies about 18.5 mEq of alkali. There was no between-group difference in urinary free deoxypyridinoline relative to creatinine, in P1NP or in beta C-telopeptide. Deoxypyridinoline was lower in the high-dose arm at four to six weeks (p = 0.04) and that did not persist. Spine bone mineral density loss was 1.8 plus or minus 3.9 percent on placebo against 2.1 plus or minus 3.2 percent across the treatment groups (p = 0.88). The authors concluded that alkali provision does not explain any long-term benefit of fruit and vegetables on bone. Jehle 2006 goes the other way and is a good trial. 161 postmenopausal women with a T score between minus 1 and minus 4 were randomised double blind to 30 mEq/day of potassium citrate or 30 mEq/day of potassium chloride for twelve months. The intergroup increase in bone mineral density was 1.87 plus or minus 0.50 percent at L2 to L4, 1.39 plus or minus 0.48 percent at the femoral neck and 1.98 plus or minus 0.51 percent at total hip, all p < 0.001, with sustained reductions in urinary calcium, rising urinary citrate indicating sustained systemic alkalisation, and reduced bone resorption markers. Distal radius and total body were not significant. Because both arms received potassium, the comparison isolates the citrate anion, which makes it the cleanest available test of the alkali claim, and it is positive. Reconciling the two is not difficult and is worth doing plainly: a potassium salt given at pharmacological dose to women with established low bone mass is a different intervention from eating fruit instead of cheese, and the trial that tested the food version alongside the salt version found both null. On the cancer claim there is almost nothing to reconcile. A systematic review screened 8,278 citations and reviewed 252 abstracts; one study met the inclusion criteria, no randomised trials were located, and no studies at all were found on dietary acid load or alkaline water for cancer treatment. The single included cohort found no association between diet acid load and bladder cancer, at an odds ratio of 1.15 (0.86 to 1.55, p = 0.36), and 1.72 (0.96 to 3.10, p = 0.08) among long-term smokers.

Protein, and whether it confounds the result

Protein is the food group the hypothesis indicts, and this is where the failed replication matters most in practice. The calcium balance meta-analyses found no evidence that increasing the diet acid load promotes skeletal mineral loss, and the Hill's criteria review found randomised studies providing no evidence for an adverse role of phosphate, milk or grain foods in osteoporosis. The clinical review of this exact question states the position in the opposite direction: in the elderly, inadequate protein intake is a greater problem for bone health than protein excess. So anyone restricting protein specifically to lower their dietary acid load is acting on the half of the mechanism that did not survive testing, and may be acting against the half of the protein literature that concerns them most. The measured protein intakes and what each produced are in the protein section of this site, and the position on older adults is in the high protein entry.

The measured intakes behind the protein figures are on the protein page.

What reliably moves

Marker	Direction	From
Urine pH	Changes reliably and substantially	1.02 units between matched acid-forming and base-forming diets. The calculated potential renal acid load of foods correlates with urine pH at r = 0.83, so the food classification predicts this marker well.
Blood pH	Barely	An average difference of 0.014 units between deliberately acid-forming and base-forming matched diets, statistically significant at p = 0.002 and physiologically trivial. This is the number that decides what the cancer claim can and cannot rest on.
Urine calcium	Down on an alkaline diet, reliably	74 percent lower on the base-forming diet in the matched crossover, and about 66 mg/day per 47 mEq change in net acid excretion in the pooled analysis. Consistently replicated and consistently misread as bone loss.
Whole body calcium balance	Unchanged	No relationship between change in net acid excretion and change in calcium balance, p = 0.38 at 94 percent power, across the five studies meeting the strictest methodological standard. The calciuria does not reflect a net loss of body calcium.
Bone resorption markers	Disputed	N-telopeptides showed no relationship with net acid excretion (p = 0.95) in the balance meta-analysis; C-telopeptide ran 19 percent higher on the acid-forming diet in the eight-person crossover; deoxypyridinoline fell with potassium citrate in one randomised trial and not in another. The marker literature does not agree with itself.
Bone mineral density	Null on food, positive in one potassium citrate trial	No difference at two years in a 276-woman four-arm trial including a fruit and vegetable arm (p = 0.88 at the spine). Plus 1.87 percent at the lumbar spine at twelve months in 161 women on potassium citrate against potassium chloride.
Fractures	Never measured	No intervention study has provided direct evidence of osteoporosis progression, meaning a fragility fracture or bone strength by biopsy. The outcome the whole hypothesis is about has not been an endpoint in any trial of it.
Cancer incidence or treatment response	One cohort, null, and nothing else	After screening 8,278 citations, one bladder cancer cohort with an odds ratio of 1.15 (0.86 to 1.55). No randomised trials located and no studies at all of alkaline diet or alkaline water as cancer treatment.

Long term, and hard outcomes

Two years is the longest trial and it is null on food. No trial of this pattern has measured a fracture, and no trial has measured a cancer outcome. The most useful long-term statement available is the list of what the causal assessment found missing: no intervention evidence of osteoporosis progression, no biological mechanism functioning at physiological pH across 19 cell studies, and prospective cohorts uncontrolled for weight loss during follow-up, family history, baseline bone density and oestrogen status. One observation is worth adding without dressing it up as a finding. The foods this pattern pushes people toward are fruit and vegetables, and the foods it pushes them away from include processed meat and hard cheese, so the diet as practised is unlikely to do harm even though the reason given for it did not replicate. That is not the same as the mechanism being correct, and the two are constantly conflated in the material that sells it.

Citations

Review2011
Causal assessment of dietary acid load and bone disease: a systematic review & meta-analysis applying Hill's epidemiologic criteria for causality
Nutrition Journal
55 of 238 studies, including 22 randomised interventions, 11 prospective cohorts and 19 cell studies. Urine calcium was consistent with the hypothesis and calcium balance was not. No intervention study provided direct evidence of osteoporosis progression, no study revealed a mechanism at physiological pH, and cohorts were uncontrolled for weight loss, family history, baseline bone density and oestrogen status.
Review2009
Meta-analysis of the effect of the acid-ash hypothesis of osteoporosis on calcium balance
Journal of Bone and Mineral Research
Five of 16 studies met the strictest methodological standards for calcium balance. Significant linear relationship between rising net acid excretion and urine calcium (p < 0.0001); no relationship with change in calcium balance (p = 0.38 at 94 percent power) and none with N-telopeptides (p = 0.95). Changes in urine calcium do not accurately represent calcium balance.
Review2009
Phosphate decreases urine calcium and increases calcium balance: a meta-analysis of the osteoporosis acid-ash diet hypothesis
Nutrition Journal
12 studies, 30 intervention arms, 269 subjects. Higher phosphate intakes decreased urine calcium excretion and increased calcium retention at both high and low calcium intake and regardless of protonation. Every finding ran contrary to the acid-ash hypothesis.
Review2008
Meta-analysis of the quantity of calcium excretion associated with the net acid excretion of the modern diet under the acid-ash diet hypothesis
American Journal of Clinical Nutrition
25 of 105 studies. Net acid excretion on control diets averaged 47 mEq/day, and a 47 mEq change was associated with 1.6 mmol/day (66 mg/day) of urine calcium in acidic urine, with a significant linear relation in acidic and alkaline urine. The authors state this is not evidence that the excreted calcium came from bone.
Human2001
Diet acids and alkalis influence calcium retention in bone
Osteoporosis International
Eight volunteers, four-period double crossover, acid-forming against base-forming diets matched for calcium, phosphate, sodium, protein and calories. Blood pH differed by 0.014 (p = 0.002) and urine pH by 1.02 (p < 0.0001). Urine calcium 74 percent higher and C-telopeptide 19 percent higher on the acid-forming diet. The trial that supports the mechanism, at n = 8.
Human2008
Effect of potassium citrate supplementation or increased fruit and vegetable intake on bone metabolism in healthy postmenopausal women: a randomized controlled trial
American Journal of Clinical Nutrition
276 postmenopausal women, two years, four arms including 300 g/day extra fruit and vegetables. No between-group difference in bone turnover markers. Spine bone density loss 1.8 percent on placebo against 2.1 percent on treatment (p = 0.88). Authors conclude alkali provision does not explain any long-term benefit of fruit and vegetables on bone.
Human2006
Partial neutralization of the acidogenic Western diet with potassium citrate increases bone mass in postmenopausal women with osteopenia
Journal of the American Society of Nephrology
The positive trial. 161 postmenopausal women with T score minus 1 to minus 4, randomised double blind to 30 mEq potassium citrate or 30 mEq potassium chloride for twelve months. Intergroup bone density increase 1.87 percent at L2 to L4, 1.39 percent at femoral neck and 1.98 percent at total hip, all p < 0.001. Both arms received potassium, isolating the citrate anion.
Review2016
Systematic review of the association between dietary acid load, alkaline water and cancer
BMJ Open
8,278 citations screened, 252 abstracts reviewed, one study met inclusion. No randomised trials located and no studies at all of dietary acid or alkaline water for cancer treatment. The one included cohort found no association with bladder cancer (OR 1.15, 0.86 to 1.55, p = 0.36). Promotion for cancer prevention or treatment described as not justified.
Review2012
The alkaline diet: is there evidence that an alkaline pH diet benefits health?
Journal of Environmental and Public Health
The most-cited affirmative review, included here because it is what advocates cite. A narrative PubMed search across pH, potential renal acid load, bone, muscle, growth hormone, back pain, vitamin D and chemotherapy, which also evaluated lay books. Its own conclusion is that there may be some value and that further studies are warranted.
Human1995
Potential renal acid load of foods and its influence on urine pH
Journal of the American Dietetic Association
The classification underlying every study in this block. Calculated potential renal acid load ranges from an average maximum of 23.6 mEq per 100 g for some hard cheeses through 0 for fats and oils to about minus 3 for fruits, juices and vegetables, and predicted net acid excretion correlates with urine pH at r = 0.83 (p < 0.001).
Review2013
Does a high dietary acid content cause bone loss, and can bone loss be prevented with an alkaline diet?
Journal of Clinical Densitometry
Clinical review. Acid loading or a high protein intake may raise urine calcium, and the evidence supporting a role in the development of osteoporosis is not consistent. Intervention studies with a more alkaline diet or with potassium citrate or bicarbonate have not consistently shown a bone benefit. States that in the elderly, inadequate protein intake is a greater problem for bone health than protein excess.

What people report

These are uncontrolled self-reports, not evidence. They are here because they tell you what to expect and what to watch for, which the trial literature does not. They cannot tell you whether anything works.

More energy and less bloating in the first weeks, reported consistently and equally consistent with simply eating more fruit and vegetables and less processed food.
Urine pH strips turning the expected colour within a day or two, which is the one effect the physiology clearly predicts and which is routinely read as evidence of a systemic change it does not demonstrate.
No change in any blood marker on retest, reported often by people who tested, and the expected result given what the crossover measured.
Reduced joint pain, reported frequently in the community and never measured as an outcome in any trial of this pattern.
Difficulty and expense, particularly around alkaline water systems, the most common reason people stop.
Anxiety about acidity, described by some who left the pattern as the part that outlasted the diet.
Unintentionally low protein intake, reported by people who cut meat, fish, eggs and dairy because of the acid load score, which is the specific failure mode the bone literature warns about in older adults.

Sources: Alkaline diet communities, wellness publishing and the commercial alkaline water sector, which has a direct financial interest in the claim. Note that the most-cited affirmative review in the medical literature explicitly evaluated lay books alongside published studies and concluded only that there may be some value and that further studies are warranted. Uncontrolled self-report throughout.

Who this is wrong for

Anyone using it instead of treatment for a diagnosed cancer. A systematic review screened 8,278 citations and found no trial of alkaline diet or alkaline water for cancer treatment of any kind, and no randomised trial of either for cancer prevention.
Anyone with kidney disease, adrenal insufficiency, or on a potassium-sparing diuretic, an ACE inhibitor or an angiotensin receptor blocker. The alkali trials delivered 18.5 to 55.5 mEq/day of potassium salts under medical supervision, and potassium load is a prescriber's question rather than a forum's.
Anyone restricting protein to lower their acid load, and older adults in particular. The clinical review of this exact question concludes that in the elderly inadequate protein intake is a greater problem for bone health than protein excess, and the balance studies found no evidence that dietary acid load causes skeletal mineral loss.
Anyone expecting their blood pH to change. The measured difference between deliberately acid-forming and base-forming matched diets was 0.014 units, against a urine pH difference of 1.02.
Anyone paying a premium for alkaline water on the strength of the cancer claim, or being sold a urine pH test strip as a measure of systemic acid-base status.
Anyone with a history of disordered eating, for whom a pattern that sorts every food onto a good or bad side of a calculated score is a poor structural fit.

Questions

Does an alkaline diet change your blood pH?: By about 0.014 units. That figure comes from a four-period double crossover in which the same eight people ate deliberately acid-forming and base-forming diets matched for calcium, phosphate, sodium, protein and calories. Urine pH in the same study differed by 1.02 units. Urine pH is what the diet changes; blood pH is what the cancer claim needs, and the difference between those two numbers is the whole argument.
Does it protect your bones?: The first half of the mechanism replicates and the second does not. Raising the diet acid load reliably raises urine calcium, at about 66 mg a day per 47 mEq change. But across the five studies meeting the strictest methodological standard, there was no relationship between change in net acid excretion and change in whole body calcium balance, at p = 0.38 with 94 percent power. The calcium in the urine does not appear to be a net loss from the body.
What about the trial where potassium citrate increased bone density?: It is real and it belongs in any honest account. 161 postmenopausal women with low bone mass were randomised to potassium citrate or potassium chloride for twelve months, and the citrate arm gained 1.87 percent at the lumbar spine, 1.39 at the femoral neck and 1.98 at total hip, all p < 0.001. Both arms got potassium, so the comparison isolates the alkali. The complication is that a separate two-year trial tested potassium citrate at two doses alongside 300 g a day of extra fruit and vegetables and found no difference in bone density or turnover from placebo.
Can it treat cancer?: There is no trial. A systematic review screened 8,278 citations and reviewed 252 abstracts. One study met the inclusion criteria, no randomised trials were located, and no studies of any kind were found on dietary acid load or alkaline water as a cancer treatment. The one included cohort found no association between diet acid load and bladder cancer. That is not a demonstration that the claim is false; it is a demonstration that after an exhaustive search there is essentially nothing to evaluate.
Should I cut protein to lower my acid load?: The evidence that would justify that is the evidence that failed. The balance meta-analyses found no sign that dietary acid load causes skeletal mineral loss, and the causal assessment found randomised studies providing no evidence for an adverse role of phosphate, milk or grain foods in osteoporosis. The clinical review of this specific question states the opposite concern: in the elderly, inadequate protein intake is a greater problem for bone health than protein excess.