Natural Oils for Baby Skin and Massage: What Research Shows

Parents are often handed oil and told it is good for a baby. Grandparents may swear by one oil, a friend by another, and a shop shelf by a third. This page looks at what the research actually shows, in plain words. The honest short answer is that the studies are fewer and more mixed than the advice, and that the answer depends a lot on the baby: a healthy full-term baby at home is a very different case from a very small premature baby in a hospital.

This page is information, not medical advice. Your midwife, health visitor or pediatrician knows your baby.

Read this first

• Do not give oil by mouth or put it in a baby's mouth or nose. A case series of 12 infants in South Africa described serious lung illness (lipoid pneumonia) after repeated oil by mouth, given for colic, constipation or a blocked nose. [23]
• Olive oil is the oil with the clearest warning. A trial in adults found it weakened the skin barrier, and a small newborn trial found it did not help the skin's fat layers. The NHS says not to use it for cradle cap. [1][2][15]
• Peanut (arachis) oil should be avoided on baby skin. The NHS says so for cradle cap, and a UK study linked peanut oil in skin products to later peanut allergy. That study cannot prove cause. [14][15]
• Moisturizing a healthy baby will not prevent eczema. Two large trials and a Cochrane review found no protection, and the review found signs of more skin infections. [10][11][13]
• The NHS says to wait until a baby is at least 1 month old before using oils or lotions. [16]

Two very different babies

Most trials in this field fall into one of two groups, and the results do not carry from one to the other.

Do not use trials of the first group to decide what to put on a healthy newborn at home.

Premature babies: oil in the hospital

Bangladesh, 2005 (a randomized trial in a hospital): 497 preterm babies born before 33 weeks were massaged with sunflower seed oil, a petroleum-based ointment, or nothing. Babies given sunflower oil were 41% less likely to get a hospital-acquired infection than the untreated babies. The ointment group showed a similar size of effect that did not quite reach statistical significance. No adverse events were seen. [4]

A Cochrane review (2021) pooled 22 trials with 5,578 preterm infants. Plant oils may reduce serious infections (RR 0.71), but the authors rated the certainty of the evidence as low, and there was little or no clear effect on deaths. Ointments and creams showed little or no effect on infection. [5] A later meta-analysis of 11 trials of oil massage found 21% lower infection risk, and the authors said more studies were needed, especially from Africa. [9]

Weight gain: a 2024 meta-analysis of 18 studies and 1,454 preterm infants found babies given oil massage gained more weight, about 52 g more on average. Results varied enormously between studies, so the average is rough. [8]

In a community setting, the picture is less clear. In a trial of 26,587 newborns in Uttar Pradesh, India, families were advised to use sunflower seed oil; the comparison group used their usual oil, typically mustard. There was no overall difference in newborn deaths. Only about 3 in 10 families in the sunflower group used sunflower oil exclusively. Among babies under 1,500 g the death rate was lower with sunflower oil. The trial was not blinded and relied on what caregivers reported. [7]

Healthy full-term babies: the barrier studies

The skin's outer layer is a barrier that keeps water in and germs out. Several studies asked whether oils help or harm it.

What this means: mustard oil massage is traditional in parts of South Asia. The lab evidence on mustard oil comes from mice, and the two large community trials we read did not show that switching to sunflower oil clearly saves lives for healthy full-term babies. We did not find a source we read showing mustard oil harms a healthy baby's skin in a trial of babies. That is not the same as showing it is safe.

A review of 16 randomized trials of emollients (oils, creams and ointments) in newborns concluded that for healthy full-term babies, emollients may not prevent eczema (RR 1.29, low certainty). [24]

Will oil or moisturizer prevent eczema?

Many parents hear that daily moisturizer in the first year stops eczema. The best evidence says it does not, for babies in general or for babies with a family history.

• BEEP trial (UK, 2020): 1,394 newborns at higher risk of eczema got daily emollient (a cream or gel) plus usual advice, or usual advice alone, for a year. At age 2, eczema was 23% versus 25% (no real difference). Skin infections were more common with emollient (IRR 1.55). The authors found no evidence that it prevents eczema in high-risk children. [10]
• PreventADALL trial (Norway and Sweden, 2020): by 12 months, atopic dermatitis occurred in 8% with no intervention and 11% with skin emollients. The authors did not support emollients for prevention. [11]
• Cochrane review (2022): 33 trials, 25,827 children. Skin care showed no clear effect on eczema (RR 1.03). Skin infections probably increased (RR 1.33). [13]

These trials tested particular creams, gels and a bath oil, not every natural oil. Babies who already have eczema are a different case: the American Academy of Pediatrics (AAP) recommends moisturizer after bathing, with ointments or fragrance-free creams preferred over thin lotions, and says to see a doctor for infection signs such as oozing, crusting or pus. [21]

Peanut oil and nut oils on baby skin

In 2003, researchers studied a UK group of children (the Avon study). Among children with peanut allergy, using skin products containing peanut oil in infancy was linked to a higher chance of peanut allergy (OR 6.8, but with a very wide range, 1.4 to 32.9, from parent interviews). Rashes that were oozing or crusted, and soy milk or formula, were also linked. The authors suggested allergy might begin when peanut oil is put on inflamed skin. [14] This was an observational study, so it shows an association, not proof.

What guidance says now, from sources we read:

• The NHS says not to use peanut oil on cradle cap because of allergy risk. [15]


• The AAP warns that massage oils, lotions and hair products may contain nut oils; families with nut or seed allergy should avoid them. It also says to read labels, because peanut oil can appear as "arachis oil." [20]


• Current research in this area is mostly about feeding: in PreventADALL, early introduction of allergenic foods lowered food allergy at 36 months. [12] Ask your pediatrician about when and how to introduce peanut.

See also nut and seed allergy basics and sesame oil on skin.

What health bodies say

The sources we read do not give a recommendation for or against baby massage with oil in healthy babies. The NHS says massage after a bath can help a baby relax, but to wait until 1 month before using oils. [16]

Cradle cap

Cradle cap is yellow, greasy or crusty scales on a baby's scalp. Both the NHS and AAP say it is usually harmless and does not need treatment.

• The NHS suggests massaging a moisturizer or coconut oil into the scalp to loosen scales, then brushing gently and washing with baby shampoo. It says not to use olive oil or peanut oil. [15]
• The AAP suggests, for stubborn scales, a small amount of mineral oil, coconut oil or petroleum jelly left on overnight before washing, with mild shampoo about every other day. [18]
• Do not pick crusts. See a doctor if it spreads, weeps, bleeds, looks swollen or smells, or if it has not improved after a few weeks. [15][18]

Fragrance, patch testing and bathing

• Fragrance: the AAP suggests fragrance-free lotion, wipes and diaper cream. [20] Scented "baby massage oils" often contain added fragrance; check the label. (The sources we read do not give rates of reaction to specific fragrances in babies.)
• Patch test: put a tiny amount on a small patch of skin and wait before wider use. Our guide explains how: patch testing natural oils and butters. We did not read a baby-specific patch test protocol for this page, so ask your health visitor.
• Bathing: the NHS says babies do not need a bath every day and that plain water is best in the first month; never leave a baby alone in the bath. [16][17]
• Essential oils are a different thing from the fixed oils on this page and are concentrated; see carrier oils explained. We did not read a source on essential oils for babies here, so ask a clinician before using any.

Questions to ask your pediatrician

• Is my baby's skin dry enough to need a moisturizer, and which product do you suggest?
• Does my baby have eczema, cradle cap or something else?
• My family has allergy (eczema, asthma, hay fever, food allergy). Does that change what I should use?
• Is it all right to massage my baby, and when should I start?
• My baby was born early or small. Is oil massage something our care team advises?
• My family traditionally uses a particular oil. Is it suitable?
• What signs of infection or allergy should send me back to you?

Related pages

• Carrier oils explained
• Check before you use natural oils and butters
• Patch testing natural oils and butters
• Olive oil
• Coconut oil
• Sunflower oil (food)
• Babies and infants: natural remedies to ask about