Benefits
HOMA-IR score and fasting insulin in a one-month marjoram tea trial
In a double-blind trial of 25 women with polycystic ovary syndrome, marjoram tea twice daily for one month changed HOMA-IR, a calculated insulin resistance score, significantly more than placebo tea, although the fall within the marjoram group alone was not significant (P = 0.06). Fasting insulin fell within the marjoram group, but not significantly more than on placebo (P = 0.08).
DHEA-S, an adrenal androgen, in the same tea trial
In the same 25-woman trial, DHEA-S fell by a mean of 1.4 micromol/L on marjoram tea, a change that compared with placebo reached P = 0.05, right at the usual significance cutoff. The abstract also lists FSH, LH, progesterone, estradiol and total testosterone as measured but reports no significant result for them. A small one-month study.
Antioxidant enzymes and protein oxidation with daily marjoram tea
In a placebo-controlled trial of 55 adults with Parkinson's disease, marjoram tea once daily for 30 days was followed by higher superoxide dismutase and catalase and lower carbonylated proteins (a marker of protein oxidation) than placebo tea. These are blood markers measured before and after, not symptoms, and the trial was small.
Non-motor symptom and mood scores alongside usual medication
In 60 adults with Parkinson's disease on their usual medicines (51 finished), 30 days of marjoram tea did not significantly change the motor score (p = 0.069). Non-motor symptom and depression questionnaire scores improved within the marjoram group and not on placebo, but the abstract does not report a direct comparison between the groups.
Hormone measures with a marjoram and cinnamon tea blend
In 60 women with polycystic ovary syndrome given a marjoram and cinnamon tea or a placebo tea twice daily for three months, only 36 finished. FSH rose within the blend group, but AMH fell in both groups by a similar amount, and the paper compared each group only with its own baseline. The blend does not show what marjoram alone does.
Glycation and platelet activity in lab and rodent studies
In lab tests, a methanol extract of marjoram leaves slowed the formation of advanced glycation end products, and in diabetic rodents 28 days of the extract improved glucose and glycation markers. A separate lab study found the extract reduced the adhesion of human platelets in a dish. Neither has been tested in people.
Mechanism of action
Essential oil chemistry differs from oregano
Marjoram herb holds roughly 2.5 to 3% essential oil. The common chemotype, which gives the typical marjoram odor, is rich in trans-sabinene hydrate and terpinen-4-ol, but at least five chemotypes have been described, and one Egyptian crop shifted between thymol, sabinene hydrate and terpinolene as main components by season.
Antiglycation activity (lab and animal data)
A methanol leaf extract inhibited advanced glycation end-product formation in vitro, partly by trapping the reactive carbonyl methylglyoxal, and protected LDL from oxidation and glycation. In streptozotocin-diabetic rodents it improved glucose and collagen glycation markers over 28 days.
Platelet adhesion and aggregation (lab data)
At 200 micrograms per mL, a methanol extract of marjoram leaves cut thrombin-activated human platelet adhesion to laminin-coated plates by 40% and also altered aggregation and protein secretion. This was done with isolated platelets in the lab; effects after drinking marjoram tea are unknown.
Clinical trials
Randomized, double-blind, placebo-controlled pilot trial of marjoram tea or placebo tea twice daily for 1 month; a later meta-analysis describes the preparation as herbal tea bags, two 250 mL cups a day. (Haj-Husein et al. 2016, J Hum Nutr Diet)
25 women with polycystic ovary syndrome in Jordan (14 marjoram, 11 placebo).
DHEA-S fell by a mean of 1.4 micromol/L and fasting insulin by 1.9 microU/mL within the marjoram group. Against placebo, the DHEA-S change reached P = 0.05 and the insulin change did not (P = 0.08). HOMA-IR did not fall significantly within the marjoram group (P = 0.06) but differed from placebo (P < 0.05). Other hormones measured had no reported significant result.
Randomized, double-blind, placebo-controlled pilot trial (PACTR202205801626909) of marjoram tea or placebo tea once daily for 30 days, with blood oxidative stress markers before and after. (Chahra et al. 2025, Neuroscience)
55 participants described as volunteers with idiopathic Parkinson's disease (30 marjoram, 25 placebo); the abstract's methods line also mentions healthy volunteers, but the trial registry entry lists only people aged 40 to 80 with Parkinson's disease as eligible.
At day 31 the marjoram group had significantly higher superoxide dismutase and catalase and lower carbonylated proteins than the placebo group, with significant time by group interactions for all three. These are blood markers; symptoms were not an outcome of this report.
Randomized, double-blind, placebo-controlled pilot study of marjoram tea for 30 days added to usual medication, scored with UPDRS III, the Non-Motor Symptoms Scale and the Beck Depression Inventory. (Chahra et al. 2021, Parkinsonism Relat Disord)
60 adults with idiopathic Parkinson's disease on conventional medication; 51 completed.
The motor score (UPDRS III) did not fall significantly (18.76 to 16.52, p = 0.069). Non-motor symptom and depression scores improved within the marjoram group (both p < 0.0001) and did not improve on placebo; the abstract does not report a direct between-group test. No side effects were detected.
Randomized, double-blind, placebo-controlled pilot trial of a powdered marjoram and cinnamon leaf tea or a placebo tea, steeped in hot water and taken twice daily for 3 months; analysis used paired before and after tests within each group, and outcomes included hormones, fasting insulin and HOMA-IR. (Majeed et al. 2025, Ann Med Surg (Lond))
60 women aged 16 to 36 with untreated polycystic ovary syndrome in Iraq; 36 completed (21 blend, 15 placebo).
FSH rose within the blend group (P = 0.032) and AMH fell within it (P = 0.003), but AMH also fell on placebo (P = 0.023), and testosterone, DHEA-S and the remaining blood measures did not change significantly in either group. No between-group test was reported. Minor complaints were mild colicky abdominal pain and a bitter taste.
Laboratory and animal safety study of a methanol extract of marjoram leaves: cytotoxicity and genotoxicity (comet and micronucleus assays) in NIH-3T3 fibroblast cells, and acute oral toxicity in rats under OECD Guideline 420. (Beceren et al. 2025, Int J Mol Sci)
NIH-3T3 fibroblast cells and rats; no human participants.
Cell viability stayed above 90% at 10 to 200 micrograms per mL, but the comet and micronucleus assays showed dose-dependent DNA damage. Rats given up to 2000 mg/kg had no deaths or overt toxicity over 14 days. The authors advise caution with concentrated formulations and call for long-term genotoxicity studies.