Cauliflower: most of what is known is about its family

Almost every number on this page was measured on the cabbage family as a whole, and where the data get specific, the vegetable measured is usually broccoli or cabbage. A pooling of 81 articles put gastrointestinal cancer risk at a rate ratio of 0.81 for the highest cruciferous intake. When the same authors went down to single vegetables, they had figures for broccoli and cabbage and none for cauliflower. The one trial that put cauliflower on a plate fed ten adults six 84 g portions of it and six of broccoli over a fortnight, and what changed was the gut, with sulphate-reducing bacteria falling in relative abundance. Even the three-fold sulforaphane advantage of light cooking, which gets quoted for cauliflower constantly, was measured in twelve people eating broccoli. None of that makes cauliflower a bad choice. It makes the evidence about it thinner than the confident advice built on top of it.

The numbers, per 100 g

How much is actually in it

• How the field was mulched more than doubled several of the plant compounds in the same variety of cauliflower. — cauliflower heads from one site under five mulching treatments, three replicates per treatment (n = 3 replicates per treatment) Evidence E source
  Combined plastic and straw mulching raised most polyphenol components, with rutin, coumaric acid, gentisic acid and caffeic acid all more than doubling against unmulched controls, and raised total flavonoids, total phenols and all four antioxidant measures

What a real portion looks like

• Six portions of cauliflower and six of broccoli over two weeks shifted which bacteria were in the gut. — 10 healthy adults; the high-Brassica diet gave six 84 g portions of broccoli, six 84 g portions of cauliflower and six 300 g portions of broccoli and sweet potato soup, against one 84 g portion of each on the low diet (n = 10) Evidence B source
  No significant change in lactobacilli (p=0.8019); the higher Brassica intake was linked to a fall in the relative abundance of sulphate-reducing bacteria (p=0.0215) and of Rikenellaceae, Ruminococcaceae, Mogibacteriaceae, Clostridium and unclassified Clostridiales (p<0.01)

What cooking and storage change

• Lightly cooked broccoli delivered about three times as much sulforaphane as fully cooked broccoli. — 12 volunteers (n = 12) Evidence B source
  Estimated yield of sulforaphane in the body was about 3-fold higher after lightly cooked broccoli than after fully cooked broccoli; the meal matrix did not significantly change hydrolysis of glucoraphanin, while pre-formed isothiocyanate from mustard was absorbed about 1.3-fold better with meat in the meal

What it does to your health

• Pooled observational data link the highest intake of cabbage family vegetables with about a fifth less colon cancer, and ten of the seventeen studies are case-control. — 17 studies, 7 cohort and 10 case-control, 97,595 patients (n = 97,595) Evidence C source
  Odds ratio 0.80 (95% CI 0.72 to 0.90) for cruciferous vegetable consumption and colon cancer, with a progressive fall in risk as intake rose in the non-linear dose-response analysis
• The same pooled data give a stomach cancer figure for broccoli and cabbage, and none at all for cauliflower. — 81 articles covering 89 studies (n = 89 studies) Evidence C source
  Highest against lowest cruciferous intake gave rate ratio 0.81 (95% CI 0.76-0.87) across gastrointestinal cancers; broccoli and cabbage specifically gave gastric cancer 0.64 (0.47-0.87) and gallbladder cancer 0.46 (0.29-0.75)

Safety, limits and interactions

• A diet loaded with cabbage family vegetables sped up one of the enzymes that clears medicines by a fifth to two fifths. — 23 dietary intervention trials using broccoli, Brussels sprouts, cabbage, cauliflower, radish or watercress with phase I and II enzyme phenotyping (n = 23 trials) Evidence A source
  Cruciferous vegetable consumption raised CYP1A2 activity by 20 to 40% and glutathione S-transferase alpha by 15 to 35%

What people believe that the data does not support

• The old warning that cabbage family vegetables harm the thyroid did not survive a review of the evidence, as long as iodine intake is adequate. — 123 studies across cell, animal and human designs (n = 123 articles) Evidence C source
  The review concludes that the vast majority of results cast doubt on the claim that brassica plants have antithyroid effects in humans, and that including them daily alongside adequate iodine poses no adverse effects on thyroid function

What is still unknown

• An umbrella review of 22 meta-analyses could say only that an association is possible, across seven different cancers. — 22 meta-analyses covering 175 independent cancer studies (n = 175 studies) Evidence C source
  Evidence on lung, gastric, prostate, breast, endometrial and ovarian cancer and renal cell carcinoma suggests a potential association with cruciferous intake; the authors call for better methods, control of influencing factors and improved evidence quality

The bottom line

If you avoid cauliflower because of your thyroid, you can stop. A review of 123 cell, animal and human studies concluded that the vast majority of results cast doubt on the old antithyroid claim, and that eating brassica vegetables daily alongside adequate iodine has no adverse effect on thyroid function. The word alongside is carrying weight there, so if your iodine intake is low, that is the thing to look at. There is one interaction worth naming out loud. Across 23 feeding trials, loading the diet with cruciferous vegetables raised the activity of CYP1A2, an enzyme that clears certain medicines, by 20 to 40 percent. If you take a drug with a narrow safety margin, a sudden jump from no cauliflower to a lot of it is worth mentioning to whoever prescribes it. On cancer, keep the claim the size the data made it. Pooled observational studies link about a fifth less colon cancer to the highest intake of cruciferous vegetables, and ten of the seventeen studies asked people to recall their diet after they were diagnosed. There it stops. An umbrella review of 22 meta-analyses covering 175 studies would go no further than calling the association across seven cancers potential, and asked for better methods and better evidence quality. One field trial explains part of why these numbers wander. Changing only the mulch under the same variety more than doubled rutin, coumaric acid, gentisic acid and caffeic acid in the heads. Two cauliflowers are not the same cauliflower.

Sources

1. clf-dr-08 · controlled field trial of growing conditions
   Among them, T1 treatment significantly increased the contents of most polyphenol components, especially the contents of rutin, coumaric acid, gentisic acid, and caffeic acid all more than doubled compared to CK1 (Fig. S3).
   BMC Plant Biol, 2026 · checked 2026-09-22
2. clf-dr-06 · randomised crossover feeding trial
   The increased consumption of Brassica was associated with a reduction in the relative abundance of SRB (p = 0.0215), and members of the Rikenellaceae, Ruminococcaceae, Mogibacteriaceae, Clostridium and unclassified Clostridiales (p < 0.01).
   Mol Nutr Food Res, 2017 · checked 2026-09-22
3. clf-dr-07 · randomised crossover trial
   The estimated yield of sulforaphane in vivo was about 3-fold higher after consumption of lightly cooked broccoli than fully cooked broccoli.
   Br J Nutr, 2007 · checked 2026-09-22
4. clf-dr-03 · dose-response meta-analysis of observational studies
   Consumption of CV was found to be inversely correlated with CC risk (odds ratios [OR] = 0.8; 95% confidence interval [CI] 0.72-0.90) in the pooled analysis using a random-effects model.
   BMC Gastroenterol, 2025 · checked 2026-09-22
5. clf-dr-05 · meta-analysis of observational studies
   High intake of broccoli and cabbage was associated with a decreased risk of gastric cancer (RR: 0.64; 95% CI: 0.47-0.87) and gallbladder cancer (RR: 0.46; 95% CI: 0.29-0.75).
   Crit Rev Food Sci Nutr, 2024 · checked 2026-09-22
6. clf-dr-02 · systematic review and meta-analysis of dietary intervention trials
   with consumption of Cruciferae increasing the activities of these enzymes by 20-40% and 15-35%, respectively.
   Clin Pharmacol Ther, 2020 · checked 2026-09-22
7. clf-dr-01 · systematic review of mixed evidence
   The vast majority of the results cast doubt on previous assumptions claiming that brassica plants have antithyroid effects in humans. Instead, they indicate that including brassica vegetables in the daily diet, particularly when accompanied by adequate iodine intake, poses no adverse effects on thyroid function.
   Int J Mol Sci, 2024 · checked 2026-09-22
8. clf-dr-04 · umbrella review of meta-analyses
   Overall, 22 meta-analyses involving 175 independent cancer studies were included. Evidence on lung, gastric, prostate, breast, endometrial, and ovarian cancer, as well as renal cell carcinoma, suggests a potential association between cancer and CV intake, which influences the risk of various cancers.
   J Food Sci, 2024 · checked 2026-09-22