Does bone broth really heal the gut, joints, and skin?

Bone broth is made by simmering bones and connective tissue for hours. That releases gelatin (cooked collagen), amino acids such as glycine and proline, and small amounts of minerals. How much ends up in the pot depends on the bones, the water, the acidity, and the cooking time^3,8.

What is in a cup. In a laboratory comparison, broth made to a standardized recipe held significantly less of the key collagen amino acids than a 20-gram collagen supplement, and recipes varied widely, with café-made broth highest; commercial broths held less than self-prepared ones³. Long, acidic cooking pulled more calcium and magnesium from bones, but a serving still provided under 5 percent of daily needs, so broth is not a mineral supplement⁸.

Gut. A 2025 review argues that broth's amino acids, such as glutamine and glycine, support the gut barrier, but it reasons from nutrients and calls for more research⁹. A randomized trial of one such nutrient gave glutamine, 5 grams three times a day for eight weeks, to people with post-infection irritable bowel syndrome: about 80 percent responded versus 6 percent on placebo¹⁰. That was a purified supplement, not broth.

Joints and skin. Meta-analyses of collagen supplement trials report less knee-osteoarthritis pain and better skin hydration and elasticity^4,5, though studies differed a lot and one analysis found no change in the pain subscore of a standard arthritis questionnaire¹¹. Those trials used collagen powders.

Sick days. Warm fluids are a sensible comfort. A systematic review of four small studies (342 people), mostly of chicken-based soups, found modest reductions in symptom severity and illness duration of one to two and a half days¹. Hot chicken soup briefly sped nasal mucus flow in 15 healthy volunteers², and chicken soup slowed the migration of neutrophils, a type of white blood cell, in a lab dish¹².

In your kitchen. Enjoy broth as a hydrating, savory food. See Liquid Gold for how to make it, what to eat and drink when ill for sick days, and collagen for the supplement evidence.

Broth itself. A PubMed search for 'bone broth' in titles and abstracts, limited to randomized or clinical-trial publication types, returned no records in October 2026. What exists is composition analyses, lab and animal work, case reports, and reviews. Alcock and colleagues measured six amino acids (glycine, lysine, proline, leucine, hydroxyproline, and hydroxylysine) in commercial and laboratory-prepared broths made with standardized and non-standardized recipes³. Standardized broth contained significantly less than a 20 g reference collagen dose (P = 0.003 for hydroxyproline, glycine, and proline; P = 0.004 for hydroxylysine, leucine, and lysine); batches made to one recipe were consistent, and non-standard recipes varied widely. The authors concluded broth is unlikely to be a reliable source of collagen precursors. Hsu and colleagues ran controlled extraction experiments: lowering the broth pH from 8.38 to 5.32 raised calcium extraction 17.4-fold and magnesium 15.3-fold, and cooking beyond eight hours extracted more, yet servings still supplied under 5 percent of daily requirements⁸.

Gut. Matar's 2025 narrative review concludes that broth's nutrients could support intestinal barrier function and might help maintain remission in inflammatory bowel disease, but it states that it aims to encourage further research⁹. The one randomized trial in this area tested glutamine, not broth: 106 adults with post-infection, diarrhea-predominant IBS and raised intestinal permeability completed eight weeks of glutamine 5 g three times daily or placebo. The primary endpoint (a drop of 50 points or more on the IBS Severity Scoring System) was met by 79.6 percent of the glutamine group and 5.8 percent of the placebo group, and permeability normalized only with glutamine¹⁰. The population was narrow, and the authors call for larger trials.

Joints. Simental-Mendía's update pooled 11 randomized trials (870 people) of oral collagen for knee osteoarthritis: function scores improved by 6.46 points (95% CI 3.40 to 9.52) and pain scores by 13.63 points (6.58 to 20.67), with high heterogeneity (I² of 75 and 88 percent)⁴. García-Coronado's earlier meta-analysis found a lower total WOMAC score (a standard knee-osteoarthritis questionnaire; weighted mean difference -8.00; -13.04 to -2.95) and lower visual-analog pain (-16.57; -26.24 to -6.89), but the WOMAC pain and function subscores were not significantly different¹¹. Both analyses covered collagen supplements.

Skin. Pu and colleagues pooled 26 randomized trials (1,721 people) of hydrolyzed collagen and found significantly better skin hydration and elasticity than placebo, while noting bias in many included trials⁵.

Sick days. Lucas's 2025 systematic review found four eligible studies (342 adults) of soup for acute respiratory infection; comparators varied, results needed narrative synthesis, and symptoms and illness duration improved by 1 to 2.5 days. Two studies found lower IL-6, TNF-alpha, or CRP, and none reported on absenteeism or well-being¹. In Saketkhoo's 1978 controlled trial in 15 healthy volunteers, sipping hot chicken soup raised nasal mucus velocity from 6.9 to 9.2 mm per minute within five minutes; hot water did similarly (6.2 to 8.4), values returned to baseline by 30 minutes, and cold water slowed mucus flow². Rennard's lab study found that a traditional chicken soup inhibited neutrophil migration in a concentration-dependent way, with great variation among commercial soups¹².

Safety: conflicting and cautionary findings. Monro's small blinded study of three organic chicken broths found lead at 9.5 µg/L in a broth made from skin and cartilage and 7.01 µg/L in chicken-bone broth, against 0.89 µg/L in tap water⁷. Hsu measured lead and cadmium at only a few micrograms per serving and called the risk minimal⁸. The two studies used different methods, and neither measured blood lead in people. Broth also contains preformed creatinine. After a kidney transplant patient ate soup containing 3,098 µmol/L of creatinine, the patient's serum creatinine rose from 128 to 171 µmol/L and returned to baseline the next day; in six healthy volunteers it rose from 68 to 77 µmol/L four hours after a creatinine-rich soup⁶.

Bottom line. Confidence is 1 of 5 for the headline claim. Broth's gut, joint, and skin benefits are inferred from its ingredients and from supplements, and the amounts in a cup are inconsistent. The sick-day use is low-risk and has small-study support, which makes broth a reasonable comfort food and not a treatment.