Do prunes improve bone density?

Not clearly. The first meta-analysis of prune trials, 11 trial papers with 747 people, mostly postmenopausal women with low bone density eating 1.8 to 3.5 oz (50 to 100 g) a day, found a borderline gain at the lumbar spine whose confidence interval crosses zero, with very high variation between trials and five of the papers drawn from one trial population. At the hip, the forearm and the whole body it found no difference. No trial counted fractures.

Unsettled. The pooled answer is borderline at the spine and null elsewhere, and its authors say the body of research is limited and its clinical relevance uncertain. Several single trials did find less bone loss on prunes. The trials were small, several lost more than a fifth of their participants, and none analyzed everyone as randomized.

Total hip bone density lost over a year, percent

One trial of 235 postmenopausal women. The abstract gives each figure with a spread of 0.2 percent without saying what kind, so no interval is drawn. The abstract gives no figure for the 100 g group, so it is not drawn. Source: card ev-prbd-04 below.

What the trials found

The 2026 meta-analysis reported the spine result as a standardized difference of 1.30, with a 95% confidence interval from -0.03 to 2.63 and heterogeneity of 98%. A standardized difference is not a percentage of bone, and an interval that reaches below zero means the result could be nothing. The same review found no significant difference in blood markers of bone formation or breakdown.

The largest single trial, the Prune Study, gave 235 postmenopausal women 1.8 oz (50 g) of prunes a day, 100 g a day, or none, for 12 months. Total hip density fell 1.1% in the control group and 0.3% on 50 g, a significant difference. The abstract reports this result for the 50 g group only and gives no figure for 100 g, so it does not show that the larger dose helps more. A further analysis of the same women found that cortical density and estimated strength at one site on the shin bone fell in controls and held steady across the two prune groups combined. That was a secondary outcome.

Three smaller trials point the same way, each with a gap. In 48 women aged 65 to 79 with low bone density, 50 g and 100 g a day for six months both prevented the loss of whole body density seen in controls, and the abstract does not give the size. In 160 women one to ten years past menopause, all taking 500 mg calcium and 400 IU vitamin D, 3.5 oz (100 g) of prunes a day for 12 months raised density at the spine and forearm compared with dried apple, again without a size in the abstract. The first prune trial, 58 women for three months, measured blood markers and no bone density. A bone formation marker rose against baseline on prunes, and markers of breakdown did not change.

Myth. The trial in men found nothing. In 59 men aged 55 to 80 with low or borderline bone mass, a year of 1.8 oz (50 g) or 3.5 oz (100 g) of prunes a day did not change total or spine density beyond what calcium and vitamin D gave every group.

Who should be careful

Not for everyone. The higher dose is hard to keep up. In the Prune Study 41% of women assigned 100 g a day dropped out, against 15% on 50 g and 10% of controls. The meta-analysis counted 12 of 78 women on 100 g, 15.4%, who stopped because of intolerance, against 4 of 79, 5.0%, on 50 g.

Prunes contain sorbitol, a sugar alcohol. In a double-blind test, 10 g of pure sorbitol raised gut symptoms in 20 people with irritable bowel syndrome and not in 21 healthy people. That was sorbitol itself and not prunes, but if you have irritable bowel syndrome the larger doses used in the bone trials may be hard on you.

We have no cards on prunes alongside osteoporosis medicines, in kidney disease, or in diabetes, and none of the trials tested prunes as a replacement for treatment. If you have been diagnosed with osteoporosis, prunes are not a substitute for what your doctor has prescribed.

What expert bodies say

The only health claim the European Union authorizes for prunes is that they contribute to normal bowel function. The register holds no bone claim for prunes. That means none was authorized, and it does not show the effect was tested and disproved. We found no position on prunes and bone from a US or European health body.

How we searched

Searched: a local copy of the PubMed 2026 baseline for prunes and dried plums with bone, bone density, osteoporosis and osteopenia (10 records, mostly trials), a second query for hip, bone loss and fracture (7 records) and a query for observational studies (none found). We also searched a full-text archive, which found the 2026 meta-analysis that was not yet in the PubMed copy, the EU register of health claims (one authorized prune claim, about the bowel, and three refused), a trial registry, Retraction Watch for all 8 papers used (no retractions) and the web for newer reviews. Search run on 7 October 2026.

Included: one meta-analysis, five randomized trials in bone, one further analysis of the largest trial, a sorbitol challenge study and the EU claim register.

Excluded: a narrative review tagged as a meta-analysis, analyses of the same trials that measured inflammation or other markers and no bone density, one case report, and a 2017 narrative review that leaned on animal and cell work.

What we read: abstracts for every trial, and the full text of the 2026 meta-analysis for its results, tolerance, risk of bias and conclusions.

What we could not get: the full text of the Prune Study, so its figures come from the abstract. No trial measured fractures, and we found no observational study of prune intake and bone.

What would change this answer

A larger, longer trial that analyses everyone as randomized and counts fractures.
A repeat of the 50 g hip result from the Prune Study in an independent group of women.
Trials in men and in women already taking osteoporosis medicines.

For what else prunes do and do not do, see the prunes page.

The food behind this question

Prunes: the bowel claim that passed and the bone claim that has not — numbers, evidence and safety

The same question for other foods (bone density)

Sources

ev-prbd-01 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs (PROSPERO CRD420251042329, RoB 2) · n = 747
The effect of intervention was borderline significant with a BMD that was slightly higher in the intervention group, at the lumbar spine site (SMD [95% CI] = 1.30 [−0.03, 2.63], I2 98%, p < 0.001).
Who: RCTs of prunes in humans, mostly osteopenic postmenopausal women over 50, 50 to 100 g/day
Effect: lumbar spine BMD SMD 1.30 (95% CI -0.03 to 2.63), I2 98%; no significant difference at other sites; I2 82.3 to 95.6%
Certainty: Very high heterogeneity; five papers come from one trial population; SMD, not percent change.
Nutrients 2026, Effects of Prunes on Bone Density in Humans: A Systematic Review and Meta-Analysis of Randomized Controlled Trials · checked 2026-10-07 · we read the fulltext
ev-prbd-02 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs (PROSPERO CRD420251042329, RoB 2) · n = 747
However, the overall body of research remains limited, and its clinical relevance remains uncertain.
Who: RCTs of prunes in postmenopausal women
Effect: no significant difference in bone formation or resorption markers; no trial measured fractures
Certainty: Attrition above 20% in several trials, no intention-to-treat analyses.
Nutrients 2026, Effects of Prunes on Bone Density in Humans: A Systematic Review and Meta-Analysis of Randomized Controlled Trials · checked 2026-10-07 · we read the fulltext
ev-prbd-03 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs (PROSPERO CRD420251042329, RoB 2) · n = 157
In three articles based on the same population but examining different outcomes [22,23,24], 4 of 79 participants in the 50 g prune group (5.0%) and 12 of 78 participants in the 100 g prune group (15.4%) discontinued the study due to intolerance.
Who: participants of RCTs in the 2026 meta-analysis; Prune Study arms of 79 (50 g) and 78 (100 g) women
Effect: discontinued for intolerance: 4 of 79 (5.0%) at 50 g, 12 of 78 (15.4%) at 100 g
Certainty: Tolerance reported unevenly across trials; harms not pooled.
Nutrients 2026, Effects of Prunes on Bone Density in Humans: A Systematic Review and Meta-Analysis of Randomized Controlled Trials · checked 2026-10-07 · we read the fulltext
ev-prbd-04 · Randomized controlled trial(s) · single-center parallel-arm randomized controlled trial, 12 months · n = 235
Total hip BMD decreased -1.1 ± 0.2% in the control group at 12 mo, whereas the 50-g prune group preserved BMD (-0.3 ± 0.2%) at 12 mo (P < 0.05).
Who: postmenopausal women aged 62.1 years; control (n = 78), 50 g prune (n = 79), 100 g prune (n = 78)
Effect: total hip BMD -0.3 ± 0.2% at 50 g vs -1.1 ± 0.2% in control at 12 months (P < 0.05); abstract sentence reports preservation for the 50 g group only
Certainty: Benefit reported for 50 g only; 22% dropout overall.
American Journal of Clinical Nutrition 2022, Prunes preserve hip bone mineral density in a 12-month randomized controlled trial in postmenopausal women: the Prune Study · checked 2026-10-07 · we read the abstract
ev-prbd-05 · Randomized controlled trial(s) · single-center parallel-arm randomized controlled trial, 12 months · n = 235
Dropout was 22%; however, the dropout rate was 41% for the 100-g prune group (compared with other groups: 10%, control; 15%, 50 g prune; P < 0.001).
Who: postmenopausal women in the Prune Study
Effect: dropout 41% at 100 g vs 10% control and 15% at 50 g (P < 0.001)
Certainty: Reasons not given in the abstract; the meta-analysis attributes 12 of these to intolerance.
American Journal of Clinical Nutrition 2022, Prunes preserve hip bone mineral density in a 12-month randomized controlled trial in postmenopausal women: the Prune Study · checked 2026-10-07 · we read the abstract
ev-prbd-06 · Randomized controlled trial(s) · secondary analysis of a randomized controlled trial (pQCT) · n = 235
The most notable effects were observed at the 14% diaphyseal tibia in the Pooled (50 g + 100 g) Prune group, in which group × time interactions were observed for cortical vBMD (p = 0.012) and estimated bone strength (SSI; p = 0.024); all of which decreased in the Control vs. no change in the Pooled Prune group from baseline to 12 months/post.
Who: postmenopausal women aged 62.1 years; control (n = 78), 50 g (n = 79), 100 g (n = 78)
Effect: 14% diaphyseal tibia: group x time interaction for cortical vBMD (p = 0.012) and strength index SSI (p = 0.024); pooled prune groups unchanged, controls declined
Certainty: Secondary outcome; result for the two prune doses pooled, not each dose.
Osteoporos Int, 2024 · checked 2026-10-07 · we read the abstract
ev-prbd-07 · Randomized controlled trial(s) · randomized controlled trial, 6 months · n = 48
Both doses of dried plum were able to prevent the loss of total body BMD compared with that of the control group (P < 0.05).
Who: osteopenic postmenopausal women aged 65 to 79; 50 g, 100 g dried plum or control
Effect: both doses prevented loss of total body BMD vs control (P < 0.05); TRAP-5b resorption marker fell
Certainty: Small and short; size of BMD difference not given in abstract.
Osteoporos Int, 2016 · checked 2026-10-07 · we read the abstract
ev-prbd-08 · Randomized controlled trial(s) · randomized controlled trial with active comparator, 12 months · n = 160
Dried plum significantly increased BMD of ulna and spine in comparison with dried apple.
Who: osteopenic women 1 to 10 years postmenopausal, not on HRT; 100 g dried plum or dried apple, plus 500 mg calcium and 400 IU vitamin D
Effect: BMD of ulna and spine increased significantly vs dried apple; size not given in abstract
Certainty: Dried apple control; dropout and per-protocol analysis noted by the meta-analysis.
Br J Nutr, 2011 · checked 2026-10-07 · we read the abstract
ev-prbd-09 · Randomized controlled trial(s) · randomized controlled trial, 12 months · n = 59
Results: No significant changes were observed in total bone mineral density (BMD) or lumbar spine BMD over one year.
Who: men aged 55 to 80 susceptible to or with osteopenia; 100 g, 50 g or 0 g prunes, all with 450 mg calcium and 800 IU vitamin D3
Effect: no significant change in total or lumbar spine BMD over one year; OPG and TRAP5b differences judged exploratory
Certainty: Small; all arms got calcium and vitamin D.
Nutrients, 2026 · checked 2026-10-07 · we read the abstract
ev-prbd-10 · Randomized controlled trial(s) · randomized controlled trial with active comparator, 3 months · n = 58
Serum and urinary markers of bone resorption, however, were not affected by either dietary regimen.
Who: postmenopausal women not on HRT; 100 g dried plums or 75 g dried apples daily
Effect: serum IGF-I and bone-specific alkaline phosphatase rose vs baseline with dried plums only; resorption markers unchanged; BMD not measured
Certainty: Biomarkers only, within-group change from baseline.
J Womens Health Gend Based Med, 2002 · checked 2026-10-07 · we read the abstract
ev-prbd-11 · Randomized controlled trial(s) · randomized double-blind placebo-controlled crossover challenge · n = 41
Overall gastrointestinal symptoms increased significantly after both polyols in patients with IBS only, although they were independent of malabsorption of either of the polyols.
Who: 21 healthy and 20 IBS subjects, challenges with 10 g sorbitol, mannitol or glucose
Effect: overall GI symptoms increased significantly after polyols in IBS only
Certainty: Pure sorbitol, not prunes; explains the intolerance seen at 100 g/day.
J Hum Nutr Diet, 2014 · checked 2026-10-07 · we read the abstract
ev-prbd-12 · Position of an expert body · EU register entry · n = —
POL-HC-8372 Dried plums of 'prune' cultivars (Prunus domestica L.) Dried plums/prunes contribute to normal bowel function Authorised
Who: EU food labeling
Effect: authorized: bowel function; no prune or dried plum bone claim in the register
Certainty: Absence of a bone claim means none was authorized, not that the effect was disproved.
EU Register on nutrition and health claims, claim POL-HC-8372, snapshot 2026-09-21 · checked 2026-10-07 · we read the section

How this page was made. Evidence was extracted from primary sources into cards, every number was re-checked against the source, and the text was written from those cards. Bioma Learn has no human medical reviewer at this time, and we say so rather than invent one. Methodology. This is information, not medical advice.