Ashwagandha: Benefits, Limits, and Who Should Skip It

Ashwagandha is the rare supplement that deserves both its popularity and a warning label written in plain English. Small human trials, summarized in the NIH Office of Dietary Supplements fact sheet, suggest root extracts may help support how you handle stress and may modestly improve sleep. At the same time, this is a biologically active nightshade herb that can move thyroid hormones, add to sedatives, and — rarely — injure the liver. This page holds both halves together: what it may do, at what doses, for how long the safety record actually runs, and the longer-than-usual list of people who should skip it. No cure-all language, because the evidence does not contain any.

What is ashwagandha and what is in the root?

Ashwagandha is Withania somnifera, a shrub in the nightshade family with a long history in Ayurvedic practice. Supplements use the root, sometimes root plus leaf, concentrated into extracts standardized to withanolides — the steroidal compounds researchers track as the active fraction. Two extract families dominate the trial literature and get their own page in our ashwagandha dosage guide: KSM-66, a root-only extract standardized to at least 5 percent withanolides, and Sensoril, a root-and-leaf extract standardized to at least 10 percent withanolide glycosides and therefore dosed lower.

That standardization matters more than brand stories. A capsule of plain root powder, a KSM-66 capsule, and a Sensoril capsule at the same milligram number are three different exposures. Trials used specific extracts at specific doses for specific weeks, and their results travel poorly to whatever happens to be cheapest this month. When you read any claim on this page, mentally append: in that extract, at that dose, in those people, for that long.

Question	What the sources support
Main trial uses	Stress support; sleep support
Typical extract dose	300 to 600 mg/day (KSM-66); 125 to 250 mg/day (Sensoril)
Trial length	Mostly about 8 weeks; safety described to about 3 months
Clearest signal	Modest cortisol and perceived-stress reductions in trials
Evidence gaps	Anxiety unclear per NIH; long-term safety unknown

What does the evidence say about stress and sleep?

For stress, several small randomized trials found that ashwagandha extracts reduced perceived stress scores and lowered cortisol modestly versus placebo — a 2025 meta-analysis of seven trials put the average cortisol drop at roughly 1.2 micrograms per deciliter. Read that correctly: a modest shift in a stress marker alongside better questionnaire scores, not a personality transplant. The NIH fact sheet concludes some preparations may help with stress, calls the anxiety evidence unclear — a distinction worth respecting, since anxiety disorders are medical conditions, not stressful weeks — and describes the testosterone evidence as limited. Anyone quoting you a precise testosterone percentage from ashwagandha is selling, not summarizing.

For sleep, the NIH summary is that ashwagandha might improve how well and how long you sleep, with benefits appearing small overall and strongest at doses of 600 mg a day or more taken for at least eight weeks. Our dedicated ashwagandha for sleep guide walks through that door in detail and connects it to the melatonin-free sleep stack. The mechanism most often proposed is indirect — calming the stress system rather than sedating you — plus possible GABA-receptor activity seen in lab work. Indirect mechanisms produce indirect results: somewhat easier nights for stressed people, not a sleeping pill. If your insomnia is severe, longstanding, or paired with apnea symptoms, that is a medical workup, not an extract upgrade.

How does ashwagandha affect thyroid and hormones?

This is the section most sellers skip. A few studies show ashwagandha may affect thyroid gland function — one small trial in fifty people with subclinical hypothyroidism found shifts in thyroid hormones at 600 mg a day — and case reports describe thyrotoxicosis, an overactive-thyroid state, in users. If your thyroid is underactive, overactive, autoimmune, or medicated, an herb that nudges thyroid hormones is not a wellness accessory; it is an uncontrolled variable in a controlled condition. The NIH does not recommend ashwagandha for people with thyroid disorders, and neither does this page without your treating clinician’s explicit involvement.

The hormone story extends further. Because ashwagandha may raise testosterone, the NIH cautions men with hormone-sensitive prostate cancer against it. Because it may stimulate immune activity, people with autoimmune conditions face the same official caution, alongside anyone taking immunosuppressants after transplant or for autoimmune disease. Add the pregnancy prohibition below and a pattern emerges: ashwagandha interacts with precisely the systems — thyroid, immune, reproductive — where casual experimentation carries outsized downside. Respect the pattern even when a given interaction has not happened to you yet.

What are the common effects and the rare serious one?

Day to day, side effects in trials were usually mild: upset stomach, loose stools, nausea, and drowsiness lead the NIH list. Drowsiness is not a footnote — it is the reason sedative interactions matter and the reason evening dosing suits some users and driving-while-testing-it suits nobody. Our ashwagandha side effects page gives this topic its full safety-led treatment, including what to watch for week by week.

The rare serious concern is liver injury. The NIH notes ashwagandha extracts have been linked to liver injury in some people, and the NIH LiverTox database rates it as a likely cause in reported cases. The typical pattern in case reports: onset two to twelve weeks after starting, yellowing of skin or eyes, itching, sometimes dark urine and fatigue, with most people recovering within one to four months of stopping — and rare severe outcomes described. Rare means most users will never see this. It also means jaundice on ashwagandha is never a wait-and-see symptom. Stop the supplement and get assessed the same day, and tell the clinician exactly what extract and dose you took. Bring the bottle.

How long should you take it, and when should you stop?

Anchor your expectations to the evidence window. Trials typically measured outcomes after about eight weeks, some ran twelve, and safety is described up to about three months — beyond that, the honest statement is that safety is unclear. A reasonable pattern for a healthy adult who has cleared the safety section is a defined trial: one extract, one studied dose, eight to twelve weeks, with a decision date. If stress or sleep has not measurably improved by then, the trial failed and the answer is stop, not escalate.

Stop immediately, regardless of the calendar, for yellowing skin or eyes, dark urine, severe itching, signs of thyroid overactivity such as racing heart, tremor, and unexplained weight loss, or excessive sedation — especially if you combined it with sleep or anxiety medication. Stop before surgery, as the NIH advises against use in that window, and tell your surgical team you took it. And do not pair ashwagandha casually with its sibling in the pair literature without reading our ashwagandha and magnesium guide first.

How do you run a fair eight-week trial of ashwagandha?

If you clear the safety section and decide to try it, run the trial the way the studies were run, so your verdict at the end means something. Pick one named extract — KSM-66 at 600 mg a day in two divided doses is the most-studied pattern — and record a one-week baseline first: a daily stress rating, your sleep quality in a line, and any symptoms you already have, so later changes have something honest to be compared against. Keep caffeine, alcohol, and bedtime as steady as life allows during the trial. Changing three habits and one herb in the same month produces a story, not a result.

During the eight to twelve weeks, the checkpoints are safety-shaped, not hype-shaped. Note stomach upset or drowsiness in week one and adjust food and timing before adjusting dose. Watch, across weeks two to twelve, for the liver and thyroid signals detailed on the side-effects page: itching with any yellowing, dark urine, a racing heart, tremor, or wired insomnia that was not there before. Any of those ends the trial immediately, whatever week it is. At the end, compare against your baseline in writing. A modest, real improvement in stress resilience or sleep is a success worth keeping — for a defined period, with the three-month evidence boundary in mind. A vague feeling that it must be doing something is not data. Stop, save the money, and put the effort into the sleep and workload fundamentals that outperform every adaptogen on record.

Who should not take this?

Do not take ashwagandha if you are pregnant — it may be unsafe, with abortifacient concern in the safety literature — or while breastfeeding, where safety is unknown. Do not take it if you have a thyroid disorder or autoimmune condition, hormone-sensitive prostate cancer, or liver disease without your treating clinician’s direct approval, and do not take it in the run-up to surgery. Talk to your doctor or pharmacist first if you take thyroid medication, sedatives or anti-seizure drugs, immunosuppressants, or diabetes or blood-pressure medicines, since ashwagandha may interact with all of them per the NIH fact sheet. It is not for children or under-18s outside medical advice, and it does not treat anxiety disorders, thyroid disease, insomnia, or low testosterone. Talk to your doctor or pharmacist about your own situation before starting.