Do Barbara O'Neill's natural remedies actually work?

Her teaching is food-first. It rests on the idea that the body heals itself when it gets the right conditions: whole plant foods, water, sunlight and sleep, plus water treatments such as hot-and-cold showers and soaks, and kitchen-cupboard poultices and packs. It helps to sort the advice into what holds up, what is thin or traditional, and what a regulator has acted on.

What holds up. Whole grains instead of white flour went with lower rates of heart disease, cancer death and early death across 45 long-term studies¹. Plant-rich eating, less added sugar, enough sleep and daily daylight are standard advice too (see plant-based diets, added sugar, sleep and sunlight). Honey probably eases cough in children over one year old², ginger may ease nausea¹¹, and aloe vera gel may speed the healing of burns¹².

What is thin or traditional. Castor-oil packs, charcoal poultices, onion and potato poultices, hot-and-cold showers and Epsom-salt soaks are folk comfort measures. Warmth can soothe, but the research is small. One before-and-after study found castor-oil packs eased straining without changing how often people went³, and we found no trials showing that charcoal or onion draws out venom, infection or toxins. Our articles on poultices and compresses, castor oil packs, activated charcoal, hydrotherapy and Epsom salt baths cover each.

The regulatory record. The NSW Health Care Complaints Commission (HCCC) describes O'Neill as an unregistered practitioner who provides services as a naturopath, nutritionist and health educator. On 24 September 2019 it found that she had breached five clauses of the Code of Conduct for Unregistered Health Practitioners and posed a risk to the health and safety of the public, and it made a permanent order prohibiting her from providing any health service, paid or voluntary⁶. Its statement quoted comments from her publications that it called non-evidence-based: that raw goat's milk can replace breast milk, that cancer is a fungus treatable with bicarbonate soda, that pregnant women with group B strep need not take antibiotics, and that there are no safe vaccines⁶. It also found that she discourages mainstream treatment for cancer, antibiotics and vaccinations⁶. The HCCC's register still listed the order as permanent when we checked on 21 October 2026¹³. Orders can change, so check the register for the current position.

Using this page. The grades below apply to the uses she describes. They draw on public lecture listings, transcripts, her online shop and regulator documents, and accounts of her preparations vary, so this page gives no recipes or doses. For sick-day basics see what to eat when sick and fever; for whole-food cooking, browse the recipe library.

Grades describe the use she recommends, not the substance in general. "We found no trials" means PubMed searches on 21 October 2026 found none for that use.

Strong

Whole grains instead of white flour. In a meta-analysis of 45 prospective studies, each extra 90 g a day (about three servings) went with 22% lower risk of cardiovascular disease, 15% lower cancer mortality and 17% lower all-cause mortality; refined grains showed little link¹. The data are observational. Her other diet principles, mostly plants and little refined sugar, are also well supported (plant-based diets, added sugar).

Moderate

Honey (cough, sore throat, colds; her onion-honey syrup). A Cochrane review of six trials (899 children aged 12 months to 18 years) found honey probably eased cough more than no treatment or placebo, and may match dextromethorphan; most children were treated for one night, and the reviewers saw no strong evidence for or against it². We found no trial of the onion. Honey can carry Clostridium botulinum spores: a 1979 investigation found them in 9 of 90 samples and linked honey to infant botulism, so it should not be fed to infants⁷. More in honey.

Ginger tea. A review of 109 trials found consistent support for easing nausea and vomiting in pregnancy and for effects on inflammation, metabolic and digestive outcomes; only 43 trials were high quality, and other uses were controversial¹¹. We found no trial for colds or flu. More in ginger.

Aloe vera gel (burns). A meta-analysis of nine trials (six with healing data) found aloe shortened healing of second-degree burns by about 3.8 days against other topical treatments, with no clear difference in pain or infection¹². Cool a burn under running water first; large, deep or facial burns, and burns in children, need medical care. More in aloe vera.

Sleep, sunlight and water are sound habits with support for specific benefits (sleep, sunlight); sun exposure also raises skin-cancer risk, so follow local sun-protection advice.

Weak

Castor-oil packs (constipation, 'lumps', cysts). The only clinical study we found was a before-and-after study in two Turkish rest homes with no separate comparison group described: packs did not change how often people went but eased straining, stool hardness and incomplete emptying³. We found no trials for lumps, cysts, fibroids, tumours, gallstones or bone spurs.

Potato poultice (swollen fingers and ankles, splinters, puncture wounds). The one trial we found randomised 62 patients with forearm swelling after a heart procedure through the wrist artery to a raw potato compress or a standard medicated compress; the potato group had less swelling and pain⁴. That is a narrow hospital setting with an active comparator, not a test of infected fingers, sprains or wounds.

Activated charcoal by mouth (poisoning, gas, 'food poisoning'). Hospitals use charcoal for some recent overdoses. In a crossover study of six healthy volunteers per drug, one 50 g dose cut absorption of two anti-seizure drugs sharply (lamotrigine to 58%, oxcarbazepine to 2.8% of usual exposure)¹⁰. That was an overdose-type dose, so it does not show what capsules do, but it is why people on prescription medicines should ask a pharmacist first. Suspected poisoning needs your local poisons centre. More in activated charcoal.

Contrast showers, Epsom soaks and the 'flu bomb' (a hot drink of garlic, ginger, lemon, honey and cayenne, promoted for colds and flu; recipes vary). Warm baths soothe, but the case that hot-and-cold showers or magnesium absorbed through the skin treat illness is weak (hydrotherapy, Epsom salt baths). We found no trial of the flu bomb as a mixture; its honey has the best evidence, and claims that it cures flu are unsupported. Never swallow essential oils; cayenne, garlic or ginger packs can burn skin, especially numb skin.

None

Charcoal and onion poultices (venom, stings and infections; earache, boils and chest colds). We found no human trials of either. A poultice must never delay care for a snake bite, a sting that swells the face, a spreading infection, or earache in a small child; see poultices and compresses.

Contradicted

A salt crystal before each glass of water, offered as the quickest way to hydrate and as help for high blood pressure. Trials of cutting sodium show lower blood pressure, so adding salt runs the wrong way (sea salt, blood pressure).

Claims to leave alone

The HCCC listed four examples of comments from her publications. Each concerns serious illness, and acting on them in place of care can cause harm.

'Cancer is a fungus', treated with bicarbonate soda, diet or wraps. Cancer is the growth of the body's own cells, not a fungus, and we found no trial showing that bicarbonate soda, taken or applied, treats the cancer itself. The evidence does show the cost of replacing proven treatment: in a matched cohort, 280 people with non-metastatic breast, prostate, lung or colorectal cancer who chose alternative medicine alone had 2.5 times the risk of death of 560 matched patients who had conventional treatment (hazard ratio 2.50, 95% CI 1.88 to 3.27). The study is observational, and choosing alternative medicine alone was rare⁵.

Raw goat's milk for babies. In US dairy outbreaks from 1993 to 2006, unpasteurised products caused 60% of those with known pasteurisation status and about 150 times more outbreaks and illness per unit consumed, with people under 20 disproportionately affected⁸. Babies who are not breastfed need a regulated infant formula; goat's-milk formulas exist and are a different product.

Skipping antibiotics for group B strep in pregnancy. About half of women who carry group B strep pass it to their baby, and without antibiotics in labour 1 to 2% of those babies develop early-onset infection. The American College of Obstetricians and Gynecologists recommends screening at 36 to 37 weeks and antibiotics in labour for carriers⁹. Your midwife or obstetrician can talk through your own situation.

'No safe vaccines.' A meta-analysis of five cohort studies (1.26 million children) and five case-control studies found no link between vaccination and autism (odds ratio 0.99, 95% CI 0.92 to 1.06), or with MMR, thimerosal or mercury¹⁴. The HCCC statement also named ADHD, epilepsy and cot death, which this review did not examine.

Bottom line

Confidence is 2 of 5. Her diet basics rest on large studies, and honey, ginger and aloe have modest trial support for specific uses. Most poultices, packs and soaks are untested or weakly tested comfort measures, with at most one small study each. The four claims the HCCC listed are not supported by the evidence we reviewed. Take the food-and-sleep basics, treat the rest as comfort, and keep your doctor in the loop.