Diindolylmethane (DIM): What 14 Human Studies Show

Paid links: Some links in this article go to iHerb through an affiliate program: if you buy something after clicking one, StarterScience earns a commission and you pay the same price. That commission never changes how we describe the evidence, including when the evidence is weak. This article is general information, not medical advice — talk to a qualified healthcare professional before starting any supplement.

Diindolylmethane (DIM) is one of the few supplements whose entire human evidence base fits on a single page. We found fourteen human studies. Four of them randomised more than fifty people. The largest, in 603 women, was negative. The third largest, in 64 women, was negative. The second largest moved a blood marker but not the tissue it was supposed to change. And the most impressive positive result is an interim analysis of twenty-one men in a trial of the manufacturer's own drug.

Below is every study we located and read, with the numbers each produced. This is not a complete census — trials in journals PubMed does not index would not surface in our search, and at least one such trial exists. But these are the studies with retrievable data, and together they say something quite different from what DIM bottles say.

Your stomach makes DIM out of broccoli — that is the entire premise

Cruciferous vegetables (broccoli, cabbage, Brussels sprouts, cauliflower) contain a glucosinolate called glucobrassicin. When plant tissue is damaged and the compound breaks down, one product is indole-3-carbinol (I3C). I3C is unstable in acid: in the stomach it condenses into larger molecules, the most prominent being 3,3'-diindolylmethane. The Linus Pauling Institute at Oregon State University describes the step plainly — "In the stomach, I3C molecules undergo acid-catalyzed condensation that generates a number of biologically active I3C oligomers, such as 3,3'-diindolylmethane (DIM)."

Supplements skip the vegetable and the I3C step. Almost every trial used BioResponse-DIM (BR-DIM), a formulation of d-alpha-tocopheryl acid succinate, phosphatidylcholine and silica microencapsulated in starch, developed because crystalline DIM is poorly absorbed.

The gap between food and capsule is enormous. The Oregon State group that measured DIM metabolism in humans put it in kitchen terms: matching the DIM in two 150 mg BR-DIM capsules would take about 1 kg of freeze-dried Brussels sprout powder, roughly 10.7 kg of fresh sprouts. Nobody gets a supplemental dose from dinner — which also means the safety record of eating vegetables does not transfer to the capsule.

Every human study we could find, in one table

Read the table as a whole and a pattern appears. Where DIM was measured against a biomarker — urinary estrogen metabolites, SHBG — it usually moved the biomarker. Where it was measured against something a patient would notice or a pathologist would call disease — HPV clearance, cervical lesion progression, breast density, PSA control, prostate tissue levels — it did nothing, including in the two largest randomised trials ever run on it.

The estrogen-metabolite story does not even replicate cleanly

The mechanism behind almost every DIM product is that estrogen can be hydroxylated at the 2 position or the 16α position, and a higher 2-hydroxyestrone to 16α-hydroxyestrone ratio has been associated with lower breast cancer risk. The usual sales line is that DIM raises that ratio. The honest version is that it depends on the dose, the population and, apparently, the study.

• Raised it: Thomson 2017 at 300 mg/day for a year (+3.2 vs −0.7, p<0.001, compliance above 91%) — the strongest single demonstration. Rajoria 2011 at 300 mg/day. Both Newman database analyses.
• Did not raise it: Nikitina and colleagues gave BRCA1 carriers the same 300 mg/day for four to six weeks and reported the ratio going from 2.4 to 3.0, P = 0.35 — no significant effect. Godínez-Martínez randomised 60 premenopausal women to 75 mg/day and reported that "DIM-treated subjects did not increase their EMUR at 30 day of supplementation," with only a non-significant upward trend a month after stopping.
• Nearly: Dalessandri's 19-woman pilot at 108 mg/day, +47%, p=0.059.

So the biomarker claim is real in that it has been demonstrated more than once, and unreliable in that two studies at doses people actually buy failed to reproduce it. And even where it moves, Thomson's trial is the cautionary tale: in the same women, over the same year, breast density measured by both mammography and MRI did not change. The authors' conclusion was that "further research is warranted."

The National Cancer Institute reaches the same verdict one level up: for cruciferous vegetables generally, "Studies in humans, however, have shown mixed results," and a meta-analysis it cites found no association between cruciferous vegetable intake and breast cancer risk at all.

Acne, PMS and fat loss: what has actually been tested

Acne. DIM is now one of the most common ingredients in acne supplements. A 2024 evidence review in the Journal of Clinical and Aesthetic Dermatology stated it flatly: "To date, there are no clinical studies that have evaluated DIM supplementation for the treatment of acne." What exists is one laboratory paper on acne-causing bacteria and one case report in which DIM was one of several things a patient was doing at once. The same review concluded that "the absence of FDA regulation and evidence-based data raises concerns about their safety and efficacy."

PMS. Castañon's cervical trial collected premenstrual syndrome as a secondary outcome, which makes it the only randomised, placebo-controlled data we have. Self-reported improvement occurred in 68 of 135 women on DIM and 24 of 60 on placebo — RR 1.3 (95% CI 0.9–1.8), p=0.236. Not significant.

Fat loss. Genuinely mixed. The cervical trial weighed people: mean weight change was +0.22 kg on DIM and +0.41 kg on placebo (p=0.624), and the proportion losing 2 kg or more was 35/243 versus 13/110, RR 1.2, p=0.634 — neither significant. But Godínez-Martínez's randomised trial of 75 mg/day for 30 days in 60 premenopausal women reported that "The DIM group saw a more significant decrease in body fat percentage than the placebo group (p = 0.04)". That is a real randomised finding — and a secondary one, in a 30-day study whose primary endpoint failed. It is a reason to run a proper fat-loss trial, not a reason to sell one.

DIM blocks the androgen receptor. It is not a testosterone booster

The men's-forum version of DIM is that clearing estrogen frees up testosterone. The human measurement runs the other way: in the one-year BRCA study, mean testosterone fell from 0.42 to 0.31 pmol/L (p=0.007), alongside estradiol.

The cell biology agrees. The 2003 study that characterised DIM found it "is a strong competitive inhibitor of DHT binding to the AR" and called it "the first example of a pure androgen receptor antagonist from plants." That is an anti-androgen profile.

The less-quoted sentence from that same paper matters more for safety. Taken together with the authors' earlier reports of DIM's estrogen agonist activity, the results "establish DIM as a unique bifunctional hormone disrupter." DIM does not simply switch estrogen off; in some contexts it acts like an estrogen. Anyone with a hormone-sensitive cancer — breast, endometrial, prostate — should read that as a reason to involve their oncologist rather than a reason to self-prescribe. Anything sold in the hormone-support aisle deserves that treatment, and a documented bifunctional hormone disrupter especially.

What the label says versus what you swallow

This trips up almost everyone who reads labels carefully. BioResponse-DIM is a complex, not pure DIM. In a 2021 human metabolism study published in Drug Metabolism and Disposition, participants took "2 BioResponse DIM 150-mg capsules (45.3 mg DIM/capsule)" — a capsule labelled 150 mg delivered about 45 mg of actual diindolylmethane. Other trials, including Thomson's, report the actual DIM content instead. So "150 mg" on two different bottles can mean a threefold difference in what reaches you. If you are going to compare DIM supplement labels, the number to look for is milligrams of diindolylmethane, not milligrams of the blend.

Doses in the published studies span 75 mg/day (Godínez-Martínez) to 900 mg/day (the manufacturer-run prostate trial), with 2 mg/kg/day in the cervical dysplasia trial. There is no established optimal dose, because there is no established outcome to optimise for.

Taking more is not simply more. In Reed's dose-escalation study, mean peak plasma DIM was 104 ng/mL after a single 200 mg dose and 108 ng/mL after 300 mg — the curve flattened while the side effects did not. The authors concluded that "increasing the dose to 300 mg did not result in an increase in Cmax."

Safety: the numbers behind "well tolerated"

"Well tolerated" is the phrase every DIM page borrows from Castañon's trial. It is accurate, and it is also the abstract. Table 5 of the same paper is more specific, and less flattering.

• Side effects were significantly more common on DIM. Excluding urine colour, 243 of 353 women on DIM (68.8%) reported at least one non-serious adverse event versus 99 of 166 on placebo (59.6%): RR 1.2 (95% CI 1.0–1.3), p=0.011. No single event drove it; the total did.
• Dark urine is not a rarity. It affected 110 women on DIM (31.2%) versus 16 on placebo (9.6%), RR 3.2, p<0.0001. It is harmless — coloured DIM metabolites being excreted — but roughly one user in three gets it, which is worth knowing before it happens.
• Serious events were not increased. Four (1.0%) on DIM versus six (3.0%) on placebo. Headaches were more frequent on DIM (17.8% vs 12.0%) but not significantly so (p=0.116).

The one documented serious harm, and who is most exposed to it

In the phase I prostate trial, 2 of the 4 men taking 300 mg twice daily developed grade 3 asymptomatic hyponatremia — dangerously low blood sodium — found on routine bloodwork. No symptoms; the only reason anyone knew was a scheduled blood test. That set the recommended phase II dose at 225 mg twice daily, and it was serious enough that Thomson's team excluded women with existing hyponatremia from their trial.

Low blood sodium is not equally likely in everyone. The US National Library of Medicine lists diuretic medicines ("water pills"), heart failure, kidney disease, cirrhosis and the syndrome of inappropriate antidiuretic hormone secretion (SIADH) among its causes. If any of those apply to you, the one harm documented in a controlled setting is the one you are already primed for — and you will not feel it coming. Sodium shows up on a basic metabolic panel, which is cheap to ask for.

A number, not a vibe

There is no official upper limit for DIM: no agency has set one. Working from what has been tested, here is ours. Do not exceed 300 mg/day of actual diindolylmethane without medical supervision; 150 mg/day is the better default. The reasoning: 150 mg/day is what the 603-woman trial used for six months, 300 mg/day is what the 12-month tamoxifen trial used under oncology monitoring with hyponatremia screening at entry, and 600 mg/day is where grade 3 hyponatremia appeared. Above 200 mg in a single dose, blood levels stop rising anyway — higher doses buy side effects, not exposure.

Case reports

Memorial Sloan Kettering's monograph lists central serous chorioretinopathy "in a healthy female patient after excessive daily intake of DIM for 2 months," resolving 8 weeks after stopping; a drug rash with eosinophilia and systemic symptoms; an ischemic stroke in a 38-year-old woman; and pulmonary embolism with deep vein thrombosis in a 65-year-old man. The dermatology review adds that the eye case was bilateral and that the stroke patient had taken 200 mg of DIM daily for a few months. In the thrombosis case the authors are explicit that "causality cannot be demonstrated in this single case" — that patient was 65, overweight, a former heavy smoker with a possible prior embolism. Case reports prove nothing. They are also the only adverse-event signal a supplement generates, because nobody is required to report them.

Pregnancy, breastfeeding and long-term use

Memorial Sloan Kettering's contraindication section states: "Because of DIM's potential hormonal effects, women who are pregnant, planning to get pregnant, or nursing should not take it. Women who use birth control pills should consult with a healthcare professional before taking this product." The Linus Pauling Institute puts the underlying fact more neutrally: "The safety of I3C or DIM supplements during pregnancy or lactation has not been established."

The longest published human exposure is one year. In some animal models, chronic I3C given after a carcinogen enhanced tumour development rather than suppressing it, which is why the Linus Pauling Institute reports that several experts "caution against the widespread use of I3C and DIM supplements in humans until their potential risks and benefits are better understood."

Drug interactions: what is documented, and what is not

Start with the honest baseline. The Linus Pauling Institute states: "No drug interactions with I3C or DIM supplementation in humans have been reported." Nothing below is a confirmed clinical interaction with case counts behind it — this is what the trials and the enzyme data show.

Tamoxifen — the strongest signal. In the randomised trial, women on tamoxifen who took DIM had significantly lower plasma endoxifen, the metabolite that does most of tamoxifen's work. The trial was not powered to show whether that worsens cancer outcomes, and the authors said so: further research is needed to determine whether the decrease "would attenuate the clinical benefit of tamoxifen." Combining the two on your own initiative is a bet against a drug known to work.

Menopausal hormone therapy. The 2025 Menopause analysis of 1,458 postmenopausal women on an estradiol patch found the 108 who also took DIM had significantly different urinary estrogen profiles across 6 of 10 metabolites. It is observational, and both authors work for the laboratory that sells the test — both worth weighing. Their recommendation was that clinicians should ask patients whether they are taking DIM and consider the implications for dosing.

Birth control pills. Memorial Sloan Kettering lists them under "Do Not Take if," alongside pregnancy and nursing, because "DIM may have hormone modulation effects." Ask a prescriber rather than stacking them.

Drugs cleared by CYP enzymes or the MDR1 transporter. This is the class-level warning that is actually documented. Memorial Sloan Kettering advises against DIM for people taking CYP450 substrate drugs or MDR1 substrate drugs, because "DIM may make them less effective." DIM induces CYP3A4 and MDR1 — CYP3A4 alone handles roughly 60% of therapeutic drugs — and preliminary evidence shows I3C and DIM raising CYP1A2 activity too. A 2021 study also found DIM is extensively metabolised in humans, with one hydroxylated metabolite more potent than DIM itself as an aryl hydrocarbon receptor agonist, so the pharmacology is not fully mapped. If you take anything where the blood level matters, ask a pharmacist to check it against CYP3A4 and MDR1 rather than assuming a plant compound is inert.

Acid-suppressing drugs, in the other direction. Because DIM is generated by acid-catalysed condensation, the Linus Pauling Institute notes that antacids, H2 blockers and proton-pump inhibitors "would likely prevent the generation of DIM and ICZ." Practically: on a daily PPI, the DIM you get from broccoli or from an I3C supplement is probably reduced. A pre-formed DIM capsule bypasses that step, though whether the biological effects differ has not been tested.

So should you take it?

If you are healthy and hoping for clearer skin: there are no clinical studies of DIM for acne at all, per the 2024 dermatology review. Buying it for acne is buying a hypothesis.

If you want it for PMS or fat loss: the randomised PMS data show no significant benefit; the weight data are split — one null result in 353 women over six months, one significant body-fat secondary outcome in 30 women over 30 days. That is a research lead, not a product.

If you want more testosterone: DIM is a competitive androgen receptor antagonist and the only human measurement showed testosterone falling. This is the wrong supplement.

If you are on tamoxifen, an estradiol patch, hormonal contraception, or any drug metabolised by CYP3A4: the evidence points to interference, not synergy. Do not add DIM without your prescriber.

If you are pregnant, trying to conceive, or breastfeeding: the guidance is not "be careful," it is "do not."

If you take diuretics or have heart, kidney or liver disease: the one documented serious harm is silent low blood sodium, and you are the person most predisposed to it. If you use DIM anyway, do it with a clinician and a basic metabolic panel.

If you have a specific clinical reason and a clinician involved: stay at or below 300 mg/day of actual DIM, prefer 150 mg/day, and read the panel for diindolylmethane content rather than blend weight. If you go looking at products sold on diindolylmethane's claimed benefits, treat every benefit claim on the box as unproven, because in humans none of them has been proven.

DIM is genuinely interesting: a plant molecule with a defined mechanism, real pharmacokinetic data, and trials in serious diseases. It is also, after two decades of study, a compound whose two largest randomised trials came back negative on every clinical endpoint they measured, and whose signature biomarker effect fails to replicate at supplement doses. Both things are true. A supplement label will only ever tell you the first one.

References

1. Castañon A, et al. Effect of diindolylmethane supplementation on low-grade cervical cytological abnormalities: double-blind, randomised, controlled trial. British Journal of Cancer, 2012
2. Del Priore G, et al. Oral diindolylmethane (DIM): pilot evaluation of a nonsurgical treatment for cervical dysplasia. Gynecologic Oncology, 2010
3. Thomson CA, et al. A randomized, placebo-controlled trial of diindolylmethane for breast cancer biomarker modulation in patients taking tamoxifen. Breast Cancer Research and Treatment, 2017
4. Godínez-Martínez E, et al. Effectiveness of 3,3'-Diindolylmethane Supplements on Favoring the Benign Estrogen Metabolism Pathway and Decreasing Body Fat in Premenopausal Women. Nutrition and Cancer, 2023
5. Gee JR, et al. Phase Ib placebo-controlled, tissue biomarker trial of diindolylmethane (BR-DIMNG) in patients with prostate cancer who are undergoing prostatectomy. European Journal of Cancer Prevention, 2016
6. Nikitina D, et al. The effect of oral 3,3'-diindolylmethane supplementation on the 2:16α-OHE ratio in BRCA1 mutation carriers. Familial Cancer, 2015
7. Dalessandri KM, et al. Pilot study: effect of 3,3'-diindolylmethane supplements on urinary hormone metabolites in postmenopausal women with a history of early-stage breast cancer. Nutrition and Cancer, 2004
8. Rajoria S, et al. 3,3'-diindolylmethane modulates estrogen metabolism in patients with thyroid proliferative disease: a pilot study. Thyroid, 2011
9. Yerushalmi R, et al. 3,3-Diindolylmethane (DIM): a nutritional intervention and its impact on breast density in healthy BRCA carriers. A prospective clinical trial. Carcinogenesis, 2020
10. Heath EI, et al. A phase I dose-escalation study of oral BR-DIM (BioResponse 3,3'-Diindolylmethane) in castrate-resistant, non-metastatic prostate cancer. American Journal of Translational Research, 2010
11. Paltsev M, et al. First results of the double-blind randomized placebo-controlled multicenter clinical trial of DIM-based therapy designed as personalized approach to reverse prostatic intraepithelial neoplasia (PIN). EPMA Journal, 2016
12. Reed GA, et al. Single-Dose Pharmacokinetics and Tolerability of Absorption-Enhanced 3,3'-Diindolylmethane in Healthy Subjects. Cancer Epidemiology, Biomarkers & Prevention, 2008
13. Newman M, Smeaton J. Exploring the impact of 3,3'-diindolylmethane on the urinary estrogen profile of premenopausal women. BMC Complementary Medicine and Therapies, 2024
14. Newman M, Smeaton J. The impact of 3,3'-diindolylmethane on estradiol and estrogen metabolism in postmenopausal women using a transdermal estradiol patch. Menopause, 2025
15. Vermillion Maier ML, et al. 3,3'-Diindolylmethane Exhibits Significant Metabolism after Oral Dosing in Humans. Drug Metabolism and Disposition, 2021
16. Le HT, et al. Plant-derived 3,3'-Diindolylmethane is a strong androgen antagonist in human prostate cancer cells. Journal of Biological Chemistry, 2003
17. Cruciferous Vegetables and Cancer Prevention. National Cancer Institute, National Institutes of Health
18. Indole-3-Carbinol. Micronutrient Information Center, Linus Pauling Institute, Oregon State University
19. Diindolylmethane. About Herbs, Memorial Sloan Kettering Cancer Center
20. Vaidya T, Hoffman L, Chapas A. Evaluating Common Ingredients Contained in Dietary Acne Supplements: An Evidence-Based Review. Journal of Clinical and Aesthetic Dermatology, 2024
21. Bui PV, Moualla M, Upson DJ. A Possible Association of Diindolylmethane with Pulmonary Embolism and Deep Venous Thrombosis. Case Reports in Medicine, 2016
22. Low blood sodium (hyponatremia). MedlinePlus Medical Encyclopedia, US National Library of Medicine