The short version
- A 2017 systematic review found 24 human studies of tulsi reporting clinical outcomes, and all 24 reported favourable results with no significant adverse events — a pattern more consistent with publication bias than with unusual efficacy.
- A 1998 analysis of trial abstracts by country found 75% of trials published in England favoured the test treatment, against 99% in China, 97% in Russia and the former USSR, 95% in Taiwan and 89% in Japan, with no trial from China or Russia finding a treatment ineffective.
- The strongest holy basil study is an eight-week randomised placebo-controlled trial in 100 stressed adults using 125mg of a standardised extract twice daily, which improved Perceived Stress Scale scores (p = 0.003) and lowered hair cortisol at week eight (p = 0.025).
- In that trial, participants exposed to the Maastricht Acute Stress Test showed lower salivary cortisol, lower salivary amylase, lower systolic and diastolic blood pressure and lower subjective stress ratings than placebo.
- The entire controlled case for holy basil is one trial of one branded extract at 250mg a day for eight weeks, which does not automatically transfer to tea, to a different extract, or to a multi-herb blend.
Jamshidi and Cohen, at RMIT University in Melbourne, went looking for every human study of tulsi they could find. They searched Cochrane, Medline, PubMed, Embase, Science Direct, Google Scholar, Indian medical databases, plus books, theses and conference proceedings. They found 24 studies reporting clinical outcomes, covering metabolic disorders, cardiovascular disease, immunity and neurocognition.
All 24 reported favourable clinical outcomes. None reported a significant adverse event.
That is a genuinely unusual result, and it is not the good news it looks like.
Why a perfect record is a warning
Real interventions fail sometimes. Trials are underpowered, formulations vary, populations do not respond, and somebody writes it up. When an entire literature points one way, the most likely explanation is not that the substance is unusually effective. It is that the studies which found nothing were never written up, or never published.
This has been measured. Vickers and colleagues, in Controlled Clinical Trials in 1998, classified trial abstracts by country of origin. Of trials published in England, 75% found the test treatment superior to control. The equivalent figures were 99% for China, 97% for Russia and the former USSR, 95% for Taiwan and 89% for Japan. No trial published in China or Russia found a test treatment to be ineffective. Not one.
That analysis did not examine India, and it did not examine tulsi, so we are not going to pretend it is a verdict on either. It is a demonstration that a 100% success rate in a body of literature is a known signature of publication bias rather than a known signature of a working treatment. Twenty-four out of twenty-four is that signature. The reviewers themselves close by asking for work on mechanism, dosage and which populations actually benefit — which is not what you write about a settled question.
The trial that does hold up
There is one modern study built the way you would want. Lopresti and colleagues, in Frontiers in Nutrition in 2022, ran a two-arm, parallel-group, eight-week randomised double-blind placebo-controlled trial in 100 adults aged 18 to 65 who were experiencing stress. The dose was 125mg of a standardised extract twice daily, and the trial was registered in advance.
Against placebo, the extract group showed greater improvement on the Perceived Stress Scale (p = 0.003) and on the Athens Insomnia Scale (p = 0.025), and had lower hair cortisol concentrations at week eight (p = 0.025). The more interesting half is the acute test. Participants were put through the Maastricht Acute Stress Test, a standardised laboratory stressor, and the supplemented group showed lower salivary cortisol (p = 0.001), lower salivary amylase (p = 0.001), lower systolic (p = 0.010) and diastolic (p = 0.025) blood pressure, and lower subjective stress ratings (p < 0.001) afterwards.
That is a physiological stress response measured under controlled provocation, not a questionnaire filled in at home, and it is the strongest single piece of evidence this plant has. The study was funded by the company that makes the extract, which is disclosed in the paper and is true of most nutraceutical trials — worth knowing, not disqualifying.
What this does and does not license
One well-run trial of one branded, standardised extract at 250mg a day for eight weeks in 100 people. That is the whole controlled case. It does not transfer automatically to tea, to loose leaf, to a different extract at a different concentration, or to a multi-herb blend where holy basil is one line on a panel.
We sell an Ayurvedic complex, and holy basil belongs in that tradition. We are not going to hand you the Lopresti trial and let you assume our formula is that formula. If the milligram figure for the specific herb is not on the label, or it is buried inside a proprietary blend, you cannot check it against a study — and that is true of ours or anyone's.
What we'd actually tell you
Tulsi is worth a look and it is not worth a rearranged budget. The controlled evidence is one trial deep, which is roughly where reishi sits and a good deal thinner than ashwagandha. Eight weeks at 250mg a day of a standardised extract is the protocol to copy if you want to test it on yourself. Judge it on whether difficult days feel less physically expensive, not on whether you feel transformed.
Two practical cautions. Nobody has tested continuous use much beyond eight weeks, so questions about whether to cycle adaptogens are unanswered here as everywhere. And if you are pregnant, trying to conceive, on blood-thinning medication or scheduled for surgery, this is a conversation with your doctor first — the human safety record is reassuring but it is only 24 small studies deep, and absence of reported harm in a small literature is not the same as evidence of safety.
Good questions
Is holy basil the same thing as the basil in my kitchen?
No. Culinary sweet basil is Ocimum basilicum. Holy basil is Ocimum tenuiflorum, also written Ocimum sanctum, and it is a different species with a different chemistry. Nothing in the clinical literature was done with the herb in a pasta sauce, so eating more pesto is not a smaller version of the trial.
How much tulsi should I take to match the research?
250mg a day of a standardised extract, split as 125mg twice daily, for eight weeks. That is the protocol from the one properly controlled trial. Tea is a different intervention entirely, at an unknown dose, and has not been tested against placebo for stress outcomes. If a label will not tell you the milligrams, you cannot match the study.
If every study was positive, doesn't that mean it works?
It means the opposite of what it looks like. Genuine interventions fail in some trials, and a literature with no failures in it usually reflects which studies got published rather than what the substance does. A 1998 analysis showed exactly this pattern in some national literatures, where essentially no published trial ever found a treatment ineffective.
Can I take holy basil with other herbs at the same time?
You can, and you will lose the ability to tell what did anything. Almost every trial in this field tested one plant at one dose. There is also a practical problem with blends: if the label does not state how many milligrams of holy basil are in there, you cannot check the amount against the trial, and neither can we.
Is holy basil safe to take long term?
Nobody knows, and the reassurance you will read is thinner than it sounds. The safety record consists of 24 small studies reporting no significant adverse events, and the longest controlled trial ran eight weeks. Talk to your doctor first if you are pregnant, trying to conceive, taking blood-thinning medication or heading for surgery.
Sources
- Jamshidi N, Cohen MM. The Clinical Efficacy and Safety of Tulsi in Humans: A Systematic Review of the Literature. Evid Based Complement Alternat Med, 2017. View study
- Vickers A, Goyal N, Harland R, Rees R. Do certain countries produce only positive results? A systematic review of controlled trials. Control Clin Trials, 1998. View study
- Lopresti AL, Smith SJ, Metse AP, Drummond PD. A randomized, double-blind, placebo-controlled trial investigating the effects of an Ocimum tenuiflorum (Holy Basil) extract (Holixer) on stress, mood, and sleep in adults experiencing stress. Front Nutr, 2022. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
