Do castor oil packs work for constipation, pain, or 'detox'?

Castor oil comes from the seeds of Ricinus communis, and its primary fatty acid is ricinoleic acid⁷. When swallowed, gut enzymes release ricinoleic acid, which switches on EP3 receptors on smooth muscle in the intestine and the uterus; that is why it works as a laxative and can trigger contractions⁶. A pack is different: the oil sits on the skin. We found no study showing that ricinoleic acid from a pack reaches the gut, liver, or lymph vessels in useful amounts, so the swallowed-oil mechanism cannot simply be borrowed to explain packs.

What the one study found. In two rest homes in Manisa, Turkey, older adults with long-standing constipation were monitored with symptom scales and diaries for 7 days before, 3 days during, and 4 days after pack use². Packs did not change how often people went or the amount of stool, but stool consistency, straining, and the feeling of incomplete emptying improved. The abstract describes no placebo or heat-only comparison, so oil, warmth, and attention cannot be separated.

Pain, liver, and lymph. We found no trials of packs for pain, liver 'detox', or lymph drainage: a PubMed search combining castor oil with those terms and with pack, compress, or topical terms returned nothing relevant. The one pain trial of castor oil used swallowed capsules (0.9 mL three times a day for four weeks) against the drug diclofenac in 100 people with knee osteoarthritis, with no placebo group⁸. Swallowed castor oil also increased gut permeability in a pilot trial, the opposite of 'gut healing'⁹.

Skin safety. An expert panel found castor oil is not a significant skin irritant or sensitizer in human tests, that it contains no ricin, and that it can increase skin absorption of other substances⁷; rare allergic reactions are reported³. The bigger risk with packs is heat: repeated use of heating pads on the same skin can cause erythema ab igne, a heat-related skin change, and skin cancers have occasionally been reported within such patches⁴.

If you try one. Treat it as a comfort measure beside, not instead of, fluids, fiber, and movement (see constipation). Keep any heat warm rather than hot, keep the oil off broken skin, and read about the oil itself in ricinoleic acid and castor oil.

Packs. Arslan and Eşer's 2011 study is the only clinical study of castor oil packs we found². It took place in two rest homes in Manisa, Turkey. Residents were monitored with a recall-bias visual analog questionnaire and a standard diary for 7 days before, 3 days during, and 4 days after pack use, and 80 percent had been constipated for 10 years or longer. Pack use did not change the number of bowel movements or the amount of stool, but it lowered stool-consistency scores and straining and improved the sense of complete evacuation. The abstract gives no sample size and describes no placebo, sham, or heat-only group; outcomes were self-reported and follow-up lasted four days.

Swallowed castor oil for osteoarthritis. Medhi and colleagues randomized 100 people with knee osteoarthritis, double-blind, to castor oil capsules (0.9 mL three times daily; 50 people) or diclofenac sodium (50 mg three times daily; 50 people) for four weeks⁸. Both groups improved significantly; adverse reactions were frequent with diclofenac and none were reported with castor oil. With no placebo arm, the trial cannot show the oil beat natural fluctuation, and it tested swallowed oil, not packs.

Labor induction: mixed results. Cochrane's 2013 review found three trials (233 women) of castor oil for cervical ripening or labor induction, all using single doses and of poor methodological quality⁵. Cesarean rates did not clearly differ in the two trials that reported them (risk ratio 2.04; 95% CI 0.92 to 4.55), no data on maternal or newborn harm were presented, and all women who swallowed castor oil felt nauseous (risk ratio 59.92; 8.46 to 424.52). Gilad's later trial randomized 81 low-risk, post-date women to 60 mL of castor oil (38 women) or sunflower oil (43): women who had given birth before were more likely to enter active labor within 24 hours (hazard ratio 2.93; P = 0.048), but first-time mothers were not, and no excess complications were noted¹⁰. Taken together, swallowed castor oil has a real uterine effect that is not reliably useful, which supports the pregnancy caution. In mice, the effect needs EP3 receptors on smooth muscle: animals lacking them there showed no laxative or uterine response⁶.

Gut permeability. Yang's pilot randomized trial gave healthy adults placebo or castor oil at 750, 1,500, or 3,000 mg daily and measured urinary excretion of mannitol and lactulose, markers of intestinal permeability. The 3,000 mg dose raised excretion at several time points, so castor oil may loosen the gut barrier in healthy adults; the authors present it as a way to test barrier-protective treatments, which sits poorly with 'gut healing' claims⁹.

Safety. The Cosmetic Ingredient Review concluded castor oil and ricinoleic acid are safe as used in cosmetics: animal tests showed no significant irritation or sensitization potential, and human tests found castor oil was not a significant irritant, sensitizer, or photosensitizer, although people with certain occupational dermatoses may react⁷. A 2026 narrative review calls contact dermatitis rare and asks for clinical studies³. Erythema ab igne from heating pads is preventable⁴.

Gaps and null findings. No trial has compared a castor oil pack with a plain warm cloth, so the oil's own contribution is unknown. The one pack study found no change in how often people went. We found no controlled study of packs for pain, liver, lymph, or 'detox', and animal findings on ricinoleic acid's anti-inflammatory effects do not tell us what happens when oil sits on a person's skin.

Bottom line. Confidence is 1 of 5. Packs have one clinical study with no comparison group described in its abstract, showing relief of some constipation symptoms; liver and lymph claims are untested; and the better-studied effects of castor oil come from swallowing it, which is a different intervention with its own risks.