Two foods on the usual list have real human studies behind them. Across the whole research literature, 507 plants have been tested against parasites and three — three — have been through a randomised trial in people. Here is what each food actually has, what the strongest study quietly admits, and the famous trial that was retracted.
No food reliably clears an intestinal parasite — and the same goes for the herbal deworming protocols built on them. A confirmed infection is treated with a specific prescription matched to the parasite — which is why the test comes first. What follows is what the evidence really says, ranked from strongest to weakest, because "strongest" here is more interesting than the sellers admit.
Before any single food, the shape of the whole field is worth seeing, because it explains why every answer below lands the same way.
Read that again: five hundred plants examined, and three trials in actual people. Nearly four fifths of the work never left the dish. That is not a gap the marketing fills in — it is the evidence base. Everything below sits inside those numbers.
This is the strongest entry, and it has grown since most pages were written about it. There are two human studies, not one.
Nigeria, 2007. In a pilot study, 60 asymptomatic Nigerian children with parasites visible on stool microscopy were given a single 20 mL dose of either dried papaya seed in honey, or honey alone. Seven days later, stools had cleared in 23 of 30 (76.7%) on the papaya elixir against 5 of 30 (16.7%) on honey, with no harmful effects reported.
Kenya, 2018. Larger, and far more useful, because it put papaya up against the real drug. 326 children across three schools ate porridge fortified with 10 g of papaya seed daily for two months, or plain porridge, or took a single 400 mg dose of albendazole.
| Group | What they took | Ascaris egg count after 2 months |
|---|---|---|
| Papaya porridge | 10 g seed daily, 2 months | down 63.9% |
| Albendazole | one 400 mg dose | down 78.8% |
| Plain porridge | nothing | went up |
So papaya did something measurable — and a single tablet still beat two months of daily porridge. That is the comparison the supplement pages leave out.
The authors are candid about the rest. They call it a preliminary survey and a one-site cluster study — whole schools were assigned to each arm rather than individual children, so differences between the schools could account for some of the effect. They also state plainly that they could not verify the most effective dosing regimen and frequency, and funding ended the follow-up at two months.
Two independent assessments land in the same place. Science Feedback concludes there is insufficient evidence for the claim that papaya seeds can kill intestinal parasites in general. Worth noting where it and the papers differ: Science Feedback describes both studies as specifically about roundworms, while the 2007 Nigerian abstract reports clearance across several parasite types. The disagreement is about scope, not about whether the evidence is thin — on that they agree. And gastroenterologist Dr Christine Lee, for Cleveland Clinic: “There is not adequate scientific evidence at this time to declare if this is effective or safe for human consumption” — she also notes trace amounts of cyanide occur naturally on the seeds, so quantity is not harmless.
The two studies do not cover the same ground. The Kenyan one measured Ascaris — roundworm — and nothing else. The Nigerian one was broader: clearance across the various types of parasites encountered ran between 71.4% and 100%, against 0–15.4% on honey alone. Both were in children, in high-transmission settings, and neither speaks to an adult in a country where these infections are uncommon.
This page used to say pumpkin was animal-only. That was wrong, and the correction is more interesting than the error was.
Pumpkin seeds contain cucurbitin, and a 2012 community study in Sichuan, China tested them against human tapeworm. Given alone, 9 of 12 confirmed taeniasis cases (75.0%) expelled whole tapeworms — but slowly, taking about 14 hours 10 minutes on average. Twelve people is a very small group.
The striking result was the combination: 91 of 115 suspected cases (79.1%) released whole tapeworms, in a mean of roughly two hours. Which sounds like the answer, until you ask what it was combined with.
So the honest summary is narrow. Pumpkin seed alone has one small human result, against tapeworm specifically, and it works slowly. The version that works quickly leans on a Group 1 carcinogen — not a trade worth making for a worm a single tablet treats. There is also laboratory work in mice showing pumpkin seed extract cutting egg counts, strongest at 8 g/kg of bodyweight, but that is a concentrated extract dosed by weight against a rodent worm, and it does not convert into a number of seeds for a person.
Garlic is the one where the underlying science is stronger than "folklore" suggests, and still not a treatment. Of the sulphur compounds in garlic, allicin showed the highest anti-giardial activity against Giardia in the dish — and it did not stop there: in experimentally infected gerbils, allicin eliminated trophozoites.
So the sequence is a dish, then a rodent. What does not exist is the step that matters — a trial in people with a confirmed infection, at a dose anyone would actually eat. Garlic on your food is good food. It is not a course of treatment.
Search this topic long enough and you will meet a spectacular result: a trial of Artemisia annua tea — sweet wormwood — in 800 patients with schistosomiasis in the Democratic Republic of Congo, reported to clear the parasite faster than praziquantel, the standard drug.
It is flagged on PubMed as a Retracted Publication, and the journal published a retraction notice in September 2020. The paper has been withdrawn from the scientific record.
That is the whole problem in one example. A retraction is quiet and a headline is loud. If someone offers you that study, you now know what became of it — and it is a fair test of whoever is offering it.
The framing assumes you have a parasite. Most people asking don't — bloating, fatigue and irregular digestion overlap almost entirely with IBS, diet, stress and thyroid problems. The symptom comparison is here.
And if you do have one, "eating something it dislikes" isn't how you clear it. You need to know which parasite it is, because the medication differs by species.
Here is the part that gets skipped, and it is the part with real evidence behind it. Food genuinely affects your parasite risk — just not by killing anything. It affects it at the point the food is handled.
And the handling standard is more exact than “cook it properly” suggests. For fish that will be served raw — sushi, ceviche, crudo — the law does not ask for care. It specifies numbers.
Two things fall out of that, and both are worth knowing.
The first is reassuring: the sushi you order has almost certainly been frozen. People assume fresh-never-frozen is the premium option. For raw fish it is the opposite — freezing is the safety step, and a restaurant serving raw fish is required to have done it.
The second is a caution, and it comes with a number. The FDA tells households to keep the freezer at 0°F (-18°C) — which is four degrees warmer than the -4°F the parasite rule requires. So a freezer running exactly as the FDA recommends still sits above the parasite-destruction threshold.
That is not a warning about your kitchen; it is the difference between two separate standards. One is written to keep stored food safe and stop bacteria growing. The other is written to kill parasites in fish that will never be cooked. Buying fish and freezing it at home meets the first and not the second.
None of it is exciting, which is exactly why it does not trend. It is also the only part of this page that reliably works. The full checklist is here.
The same question, one step further — and the same honest answer.
Read →Which symptoms point somewhere, and which point nowhere at all.
Symptoms guide →Is It a Parasite? — four pages, plain English, from CDC and WHO sources. The red flags and a printable prevention checklist.
