Mind & Body · Tuesday, 1 September 2026

01 Briefing what happened

Your outer skin is dead cells set in fat, and the gaps in it are where a food allergy begins

Mind & Body 1 min 73 sources

The outermost layer of your skin is not alive. It is flattened dead cells packed in fat - and when it leaks, the body meets food through the wrong door and learns to fight it.

204 million

people living with eczema, about 11.1% of children and 6.3% of adults

it carries the heaviest load of lost healthy life of any skin disease, and ranks 15th among all non-fatal diseases [22][23]

0.79% to 0.45%

peanut allergy in young US children, before and after the feeding advice was reversed

measured across 48 clinics and more than 120,000 children followed from birth to age three [15][16]

36.1% vs 43.0%

babies with eczema by age two, with daily moisturiser and without

the largest effect was in the babies who were not at high risk to begin with [18]

88%

of homes with a three-month-old where peanut protein was already in the dust

egg, cow's milk, wheat and dust mite protein were in all 26 homes tested [11]

The lead story — what happened

The outer layer of your skin, called the stratum corneum, is made of dead flattened cells packed into a mortar of fats. [1]
It does two jobs at once: holding water and nutrients in, and keeping microbes, allergens and ultraviolet light out. [1][3]
Those cells are built by filaggrin, a protein stored in droplets inside the living cell below. A change in acidity dissolves the droplets, the cell flattens and dies, and it becomes wall. [2]
Faulty copies of the filaggrin gene are the strongest known genetic risk for eczema. [7] A review of 20 studies found the odds up to 11 times higher in carriers of some versions. [6]
Genes shape who is at risk and who benefits. A later analysis of the peanut trial tied outcomes to three genes: the filaggrin gene plus two immune ones. [14]
The mortar matters as much as the bricks. In eczema the fats called ceramides are reduced and shifted towards shorter, weaker forms. [4]
The same shift happens with age. In 58 adults with dry, eczema-prone skin, dryness rose and water held in the skin fell decade by decade, tracking the change in those fats. [5]
A leaky wall does not just let things in. It changes where the immune system first meets them, and that turns out to decide what it does about them. [8][10]
Meet a food protein through the gut, with a meal, and the body learns tolerance. Meet the same protein through inflamed skin, and it learns to attack. That is the dual allergen idea. [8][9]
The exposure is real and it starts early. In dust from 26 homes with a three to four-month-old baby, egg, cow's milk and wheat protein were in every sample and peanut in 88%, before any of the babies had eaten solid food. [11]
So eczema in a baby is the single strongest early warning of a food allergy later. About 30% of infants with severe eczema go on to develop one. [8][21]
A review pooled 190 studies and 2.8 million people. The largest risk factors for food allergy were earlier allergic conditions, eczema, faster water loss through the skin, filaggrin gene variants and late introduction of solid foods. [13]
The theory was tested backwards, by feeding. Researchers asked why Jewish children in Britain had about ten times the peanut allergy of Israeli children of similar ancestry, who eat a peanut snack as babies. [15]
US guidance reversed in 2015 and again in 2017, advising peanut foods from four to six months for babies at high risk. [15][17]
It worked. Across 48 clinics and more than 120,000 children, peanut allergy fell from 0.79% to 0.53% after the first change in advice, and to 0.45% once the follow-up advice was counted too. [17][16]
Any food allergy fell the same way, from 1.46% to 1.02% and then to 0.93%. [17][16]
The doctor who led that count estimates the changed advice has spared at least 40,000 children a peanut allergy in a decade. [15]
Working from the same theory in the other direction has gone much worse. Trials of moisturiser from birth have shown no convincing sign that it prevents food allergy. [9][8]
It does reduce eczema, modestly. In 1,247 US babies not picked for risk, daily moisturiser from nine weeks old left 36.1% with eczema at two years against 43.0% without it. [18]
A separate review of lipid-based creams in high-risk babies found only trends towards less eczema, none of them statistically firm. [19]
What did cut allergy was treating the inflammation rather than greasing the surface, including on skin that looks clear. [9]
Inflamed skin reaches the gut directly. Immune cells in eczema skin prime cells in the intestine, and steroid creams leave two of the signals doing it untouched. [20][12]
None of this is a plan anyone can run at home. The guidelines are written for doctors, and the advice that followed was aimed at babies already judged to be at high risk. [42][15]

Who is involved

Gideon Lack

the allergy researcher whose name is on the skin-versus-gut theory, which says the skin teaches the body to attack a food and the gut teaches it to accept one [8]
David Hill

a children's allergy doctor at the Children's Hospital of Philadelphia in the US; he led the study that measured the fall in peanut allergy [15]
Eric Simpson and colleagues

US skin researchers who ran the 1,247-baby trial testing whether daily moisturiser from birth prevents eczema [18]
Filaggrin

the protein that builds the dead outer layer of skin; broken copies of its gene are the strongest known genetic risk for eczema [2][7]
Staphylococcus aureus

a bacterium that lives harmlessly on most people's skin and takes the place over in eczema, damaging the wall further [28][29]

What is pushing on this

Food met through broken skin High→

food protein sits in house dust from a baby's first months [11]

Food met through the gut Building↑

feeding advice reversed and allergy rates fell [15]

Greasing the surface Easing↓

moisturiser from birth did not prevent food allergy [9]

Treating the inflammation Building↑

controlling eczema, including clear-looking skin, cut egg allergy in trials [9]

How it unfolded

Until 2015 parents in Britain and the US were advised to keep babies away from peanut [15]
2015 a trial found the opposite: babies fed peanut early were far less likely to become allergic [15]
2015 to 2017 US guidance reversed, advising peanut foods from four to six months for high-risk babies [15][17]
2025 a count of more than 120,000 children found peanut allergy down 43% and any food allergy down 36% [15][16]
Now trials are testing early walnut and cashew, and whether treating skin inflammation prevents allergy [39][40][9]

Where this points

The next test is whether treating the inflammation in a baby's skin, rather than only greasing it, prevents food allergy - because moisturiser alone did not. [9][18]

The rest of the day

55 more stories on this beat.

Each with its own sources. None of these is a link to the story above.

02
Eczema is the heaviest skin disease

Eczema affects about 204 million people, 11.1% of children and 6.3% of adults, and ranks 15th among all non-fatal diseases worldwide for healthy life lost. [22][23]

Why it matters — It sets the scale of the wall problem the lead story describes.
03
Dermatitis cases up 39% since 1990

Global dermatitis cases reached 241 million in 2021, up 38.8% from 1990, though the rate adjusted for age fell slightly. Years of healthy life lost totalled 8.18 million. [24]

Why it matters — Population growth, not a worsening disease, is driving most of the rise.
04
The rarest kind carries most of the harm

Eczema is only 5.9% of new dermatitis cases worldwide but 75.8% of the healthy life the whole group costs. Contact dermatitis is 64.3% of cases and 22.0% of the harm. [25]

Why it matters — Counting cases and counting damage give almost opposite answers.
05
Skin disease is the sixth commonest

Skin and under-skin conditions affect 25.7% of people worldwide once infections spread by sex and skin cancers are set aside, ranking sixth of all disease groups and eighth for burden. [26]

Why it matters — It is one of the largest health problems that rarely gets counted as one.
06
Eczema is no longer only a child's disease

Onset after the age of 60 is increasingly common, and older adults now show high prevalence and high severity. [23][27]

Why it matters — The condition is being tracked across a whole life, not just infancy.
07
Old skin loses the same fats

In 58 adults with dry, eczema-prone skin, dryness rose with age and skin water content fell, alongside a shift in ceramides towards shorter chains that tracked barrier strength. [5]

Why it matters — The same defect that starts eczema in babies returns in later life for a different reason.
08
The wall is built by a phase change

Filaggrin sits in the living cell as liquid droplets. An acidity shift dissolves them, the cell flattens and dies, and becomes part of the wall. The paper reframes eczema and two other skin diseases as faults in that step. [2]

Why it matters — It moves the defect from a missing ingredient to a mistimed process.
09
Five genes behind one wall

A meta-analysis of 20 studies in European and Asian groups tied eczema to variants in five genes: FLG, SPINK5, LAMA3, HRNR and COL8A1. Some filaggrin versions carried odds up to 11.22. [6]

Why it matters — Barrier genes, not immune genes, top the list.
10
Bacteria fill the gap the gene leaves

In mice bred without filaggrin, a signalling protein called MyD88 held back the inflammation that microbes on the skin would otherwise drive. [7]

Why it matters — A weak wall does not inflame on its own; something has to move in.
11
Staph is not a bystander

Staphylococcus aureus, common on healthy skin, gains a growth advantage in eczema and is now judged a major driver of the disease rather than a passenger. [28]

Why it matters — It settles a long argument about cause and effect.
12
The toxin, and the early clue

S. aureus uses a signalling system to switch on a toxin that worsens inflammation and barrier damage. Babies colonised early by strains with a working version of that system are more likely to develop eczema. [29]

Why it matters — It gives a testable early marker before the rash appears.
13
Sterilising the skin backfires

Broad antimicrobials can deepen the imbalance they are meant to fix, pushing research towards seeding helpful bacteria instead. [29][30]

Why it matters — The target has moved from killing the intruder to restoring the neighbours.
14
Probiotic cream did nothing

Thirteen adults with mild to moderate eczema used probiotic creams at three strengths for four weeks. There was no difference from placebo in severity, water loss through the skin or quality of life. A wider review of topical probiotics found the same picture: promising mechanisms, small samples, heterogeneous results. [31][33]

Why it matters — A null result on a product category sold hard on the microbiome story.
15
Swallowed probiotics: modest, uneven

Most children's trials using multi-strain formulas reported small improvements in severity and itch, single-strain products were inconsistent, and few studies measured the skin's bacteria at all. [32]

Why it matters — The claim rests on plausibility more than on measured change.
16
Manuka oil cream matched its own base

In 118 New Zealanders with moderate to severe eczema, the score fell from 16.3 to 8.8 on the manuka cream and from 16.9 to 8.8 on the plain base cream, a difference of -0.38. [34]

Why it matters — Both arms improved a lot; the active ingredient added nothing.
17
Seawater creams work, less well than steroids

A review of 10 studies found eczema severity improved 26% to 55% with marine mineral treatments. One trial compared them head to head: 46% improvement with steroid cream against 26% with mineral bathing. [35]

Why it matters — Real effect, oversold size - the exact shape of most wellness claims.
18
Collagen drink trial reports gains

Seventy healthy adults took 1,650 mg a day of a collagen peptide or placebo for eight weeks; wrinkle depth, skin elasticity and hydration improved in the treated group. [36]

Why it matters — Sponsored functional-food trials on healthy volunteers are the weakest end of this literature.
19
The food in your face cream

Food-derived ingredients are common in moisturisers, including those used on eczema, which under the skin-route theory could teach the body to react to those foods. [10]

Why it matters — It follows directly from the mechanism and has not been settled either way.
20
The skin talks to the gut

Beyond letting protein through, inflamed skin remodels the intestine through circulating signals, microbes and nerves. About 1 in 12 US children has at least one food allergy. [12]

Why it matters — The route is not only physical; the skin changes the gut it is teaching.
21
Steroid creams leave two signals running

Steroid creams cut most inflammation but leave two signalling molecules active in the outer skin of infants. Over 16 weeks, under-fives on dupilumab plus steroids saw allergy antibody levels drop 70%; on steroids alone they rose 30%. [20]

Why it matters — Suppressing the rash is not the same as stopping what the rash is teaching.
22
Dust mite allergy shows up before age one

Of 1,793 infants tested, 96 had dust mite antibodies. Among those under 12 months, 90.9% had eczema and every one of them had a food allergy. [37]

Why it matters — The same broken wall admits more than food.
23
Two milk allergies, two skin communities

Babies with antibody-driven cow's milk allergy had a skin bacterial mix distinct from those with the other kind, despite equally severe eczema. [38]

Why it matters — Two conditions that look identical on the skin are not the same disease.
24
Walnut is next in line

A Japanese trial is testing walnut powder given at 6 to 8 months against 12 to 14 months in high-risk babies with eczema, using powder because whole nuts are a choking risk under five. [39]

Why it matters — Tree nut allergy is rising in Japan with walnut the main trigger.
25
A trial of what the mother eats

An Australian trial is recruiting 4,412 breastfeeding women, half asked to eat at least 60 peanuts and 40 cashews a week, to see whether it lowers their babies' allergy risk. [40]

Why it matters — It tests whether tolerance can be taught through milk, before solid food.
26
Doctors advise it; few babies get it

A survey of 563 US parents mapped what they actually feed infants and what stops them, against guidance recommending allergenic foods from four to six months. [41]

Why it matters — The gap between evidence and practice is where the remaining cases sit.
27
Six levers to close the gap

A July 2025 US summit named insurance coverage, food assistance for low-income mothers and infants, standardised guidelines and public education as the ways to raise early introduction. [42]

Why it matters — The obstacle is now delivery, not evidence.
28
One clinic, both specialists

The Cleveland Clinic in the US runs a joint skin and allergy clinic for babies with eczema, on the estimate that about 30% of those with severe eczema will develop a food allergy. [21]

Why it matters — The theory is changing how clinics are laid out, not just what they prescribe.
29
Peanut advice travels badly

Early peanut has a strong effect in Western countries with high peanut allergy but little in Japan, where peanut is eaten less and allergy is rarer. Egg results are mixed and cow's milk unresolved. [8]

Why it matters — A prevention that works depends on what the local diet already contains.
30
Steroid creams remain first choice

A review of guidance from 2010 to 2024 found topical steroids effective and safe at the right strength and duration, with side effects mostly local. Steroid tablets are discouraged for long-term control. [43]

Why it matters — The newer drugs have not displaced the old first step.
31
Steroid fear, measured online

Educational posts about topical steroid withdrawal on Instagram and TikTok scored below the midpoint for accuracy and safety when rated by skin doctors. Posts by doctors averaged 4.25 out of 5. [44]

Why it matters — The condition lacks agreed diagnostic criteria, which is part of why the gap is so wide.
32
Half of surveyed homes keep a steroid cream

Of 714 people surveyed in Jordan, where the creams are sold without prescription, 46.8% kept one at home and 53.1% used it daily. Nearly a third could not name a single corticosteroid cream. [45]

Why it matters — Availability without labelling produces daily use of a drug people cannot identify.
33
Confident, and unaware of the risks

Among 100 patients in southern Rajasthan, India, all prescribed a steroid cream, at least 84% said they knew the right amount, frequency and duration - and 70% did not know the risks of unsupervised use. [46]

Why it matters — Confidence and knowledge came apart in the same survey.
34
A JAK cream beat a mid-strength steroid

In a trial arm running four weeks, ruxolitinib cream cleared 75% of eczema in 56.0% of adults against 47.1% on triamcinolone. Itch relief by day two: 42.5% against 20.5%. [47]

Why it matters — The first direct sign a non-steroid cream can outperform the standard one.
35
First drug approved for hand eczema

Delgocitinib, a cream that blocks inflammation signals inside cells, is the only drug approved by the US regulator for long-running hand eczema. England's health service body recommends it after steroids fail. [48][49]

Why it matters — Hand eczema has no diagnostic code in US records, so it has been almost invisible to measurement.
36
A cream matched an injection

An indirect comparison of trial data put delgocitinib cream level with dupilumab injections for hand eczema at 16 weeks, with every result numerically favouring the cream but none statistically significant. [50]

Why it matters — If it holds, it removes an injection from a common problem.
37
A new ointment tested in China

Ivarmacitinib ointment was tested at 27 sites in China. At eight weeks, 45.1% on the higher strength reached 75% clearance against 17.9% on the base cream, with itch relief inside 48 hours. [51]

Why it matters — The non-steroid cream field is now several drugs deep.
38
Five years of dupilumab safety data

Eight trials covering more than 7,000 patient-years found more eye inflammation and injection-site reactions than placebo, and fewer serious infections. Five years of blood tests in 2,677 patients showed no meaningful drift. [52][53]

Why it matters — The long safety question on the leading eczema drug now has an answer.
39
Half could take less of it

Of 126 children whose eczema was controlled on dupilumab, 65 reduced their dose in ordinary practice. [54]

Why it matters — Dose reduction cuts cost and exposure for a drug taken for years.
40
Same drug, three different itches

Dupilumab reduced itch in eczema, in prurigo nodularis - a disease of intensely itchy lumps - and in long-running hives, across separate trials. [55]

Why it matters — Three diseases that look nothing alike share a signalling path.
41
The mood scores did not move

In 24 adults treated for 16 weeks, the change in a general wellbeing score was not statistically significant, even though skin measures improved. [56]

Why it matters — An honest null on a claim that clearer skin lifts mood.
42
A second biologic for people the first failed

Lebrikizumab was tested in an open trial in adults and teenagers who had stopped dupilumab because it did not work or they could not tolerate it. [57]

Why it matters — Non-responders were the largest gap in the treatment picture.
43
Real-world results from four countries

Dupilumab held up outside trials. Eczema scores fell from 27.6 to 2.7 over 232 weeks in South Korea. In Saudi Arabia and the UAE they fell from 49.9 to about 14 in six months. A Chinese group of 376 improved 95.3% by 16 weeks, and a Japanese extension followed children from six months old for three years. [58][59][60][70]

Why it matters — Trial results usually shrink in ordinary practice; here they largely did not.
44
What the new drugs still cannot do

A review names the remaining gaps: no stable remission, patients who do not respond, side effects to manage, and itch that persists through treatment. [61]

Why it matters — The itch, not the rash, is what patients rank first.
45
A phone programme cut relapses

615 children across 12 Chinese hospitals were randomised to a smartphone education programme or ordinary consultations. Relapse at 12 weeks was 16.6% against 24.0%; the gap faded after that. [62]

Why it matters — The cheapest intervention in this whole field, and it works briefly.
46
Baby wipes, tested against water

Forearm skin was deliberately stripped, then wiped repeatedly. Wipes with a buffered acidity and added moisturiser recovered better than plain wipes, and better than water and cloth. [63]

Why it matters — The default gentle option was not the gentlest.
47
Barrier claims outrun the tests

The market for barrier-repair products is growing while no standardised method exists to test whether they protect anything. [64]

Why it matters — A claim nobody can measure is a claim nobody can disprove.
48
Three to one to one

Skin societies recommend a 3:1:1 ratio of ceramides, cholesterol and fatty acids for repairing the barrier, a specific target most moisturisers do not state. [65]

Why it matters — It turns a vague product category into something checkable.
49
Ceramide lotion tested on one leg

A multi-centre trial had participants apply a ceramide and plant-oil lotion to one leg for four weeks, using the other leg as the control, and tracked skin fats and bacteria as well as dryness. [71]

Why it matters — Using the same person as their own control removes most of the usual confusion.
50
Rosacea is a barrier disease too

The flushing, stinging and bumps of rosacea are now traced to a damaged barrier and an over-active immune response together, not to blood vessels alone. [66]

Why it matters — The same wall failure surfaces under a different name.
51
Eczema looks different in Asia

In Asian populations two different inflammation pathways carry more of the disease, barrier gene variants differ, and pollution and fine dust worsen symptoms. [67]

Why it matters — Treatments developed on one population may not transfer whole.
52
Skincare changes the skin's bacteria

Shotgun sequencing of 37 people's skin against 48 healthy controls found the bacterial mix differed measurably between those using moisturiser alone and those using moisturiser with sunscreen. [68]

Why it matters — A daily habit nobody thinks of as a treatment is one.
53
A US minister blamed vaccines

US health secretary Robert F. Kennedy Jr told a food allergy forum he doubted that a lack of early peanut exposure caused peanut allergy, and suggested aluminium in childhood vaccines instead. Specialists said there is no evidence for it. [69]

Why it matters — It contradicts the strongest prevention result the field has, at the moment it is being rolled out.
54
No blood test grades eczema

There is still no validated laboratory marker to sit alongside the visual scoring systems doctors use, despite years of searching. [22]

Why it matters — Severity is judged by eye, which is part of why trials are hard to compare.
55
Combined therapy tested in babies

87 infants aged 3 to 12 months had moderate to severe eczema plus several allergies. A combination of a ceramide-like cream, a non-steroid anti-inflammatory and an oral supplement cut severity by more than 60% by week eight, against standard care. [73]

Why it matters — A small trial, but pointed at the youngest group where prevention still matters.
56
The tolerance side is still unsolved

Food allergy affects around 10% of people worldwide, and oral immunotherapy remains the only treatment that builds tolerance - with results that vary by person and by food. [72]

Why it matters — Preventing the allergy is going better than curing it.

02 Lesson why it matters

What a thing is taken for depends on where it turned up

The body cannot tell what a protein is, so it goes by the way in - food through the gut, an enemy through broken skin.

The twist

The immune system never decides what a thing is. It decides where the thing turned up - and the same peanut becomes dinner or an enemy depending on which way in it took.

How it works

The body cannot read a protein and decide what it is
So it judges by the circumstances of the first meeting
Arriving through the gut with a meal means food, and it learns to accept
Arriving through inflamed skin means invasion, and it learns to attack
The wall is therefore not just a barrier. It is a sorting office
Break the wall and the same protein gets filed under the wrong heading

Where you've seen this

Courts

the same fact counts as evidence or is thrown out entirely, depending on how it was obtained

Hiring

one CV reads as a strong candidate when a colleague hands it over and as noise in the online pile

Borders

identical goods are an import at one channel and smuggling at another

Rumours

the same claim lands as news from a friend and as nonsense from a stranger

The catch

Knowing how something starts does not tell you how to stop it. The skin-route theory is well supported, and the treatment that followed most directly from it - greasing the skin from birth - lowered eczema a little and did not prevent food allergy at all.

And the whole of it

Everyone in this story is doing the sensible thing from where they sit. The parents who avoided peanut were following the best advice of their day. The immune system reading the doorway is using the only clue it has. Most of us are being sorted this way too, by which entrance we happened to come in by, and almost nobody can see their own.

03 Truth what's really going on

What is really going on

One theory produced two treatments and only one of them worked - feeding babies the food cut allergy sharply, while greasing the skin from birth barely shifted eczema and did not prevent allergy at all.

Why it works on us — A true mechanism is the best sales tool there is - once a broken barrier is real, repairing it sounds like a conclusion rather than a claim that still has to be tested.

Who gains

Companies selling barrier-repair skincare — The mechanism they name is genuinely real, the market is growing, and no standardised method exists to test whether a barrier product protects anything. [64][65]
The makers of injected eczema drugs — Five years of safety data and real-world results from four countries all point the same way, for a drug people take indefinitely. [52][53][58]
The maker of delgocitinib cream — It is the only drug the US regulator has approved for long-running hand eczema, in a condition that has no US diagnostic code and therefore almost no comparative data. [48][49]
Sellers of collagen drinks and mineral creams — Every finding that the skin barrier is a fats-and-water problem creates a market for anything sold as fats and water. [36][35]

Who pays

Children born under the avoidance advice — For the years the guidance said keep peanut away, peanut allergy in US children rose from 0.4% to more than 2%. [15]
Babies whose eczema is treated only as a rash — Steroid creams clear the skin but leave two signalling molecules active in the outer layer. Over 16 weeks, allergy antibody levels rose 30% in under-fives on steroids alone. [20]
People in countries where steroid creams are sold without prescription — Of 714 people surveyed in Jordan, 46.8% keep one at home and 53.1% use it daily, while 29.5% cannot name a single corticosteroid cream. In a separate survey of 100 patients in India, 70% did not know the risks of using them unsupervised. [45][46]
People with chronic hand eczema — There is no diagnostic code for it in US records, which makes the condition, its cost and its treatments nearly invisible to measurement. [48]

What nobody knows yet

Open questions from across today’s stories — ours included.

01

Whether putting moisturiser on a baby prevents eczema at all.

The two strongest sources here disagree. A trial of 1,247 US babies found 36.1% with eczema at two years against 43.0% without, a real reduction. A review of lipid creams in high-risk infants found only trends, none of them statistically firm. [18][19]
02

Why that trial worked better in the babies who were not at high risk.

The effect was larger in the lower-risk group and larger again in homes with a dog. The researchers report both and do not explain either. [18]
03

Whether food ingredients in skin creams can cause allergy to those foods.

It follows directly from the skin-route mechanism, and common food-derived ingredients differ enormously in how often they cause allergy. Nobody has run the test. [10]
04

How much of the fall in peanut allergy the new advice actually caused.

It is a before-and-after in medical records, not a randomised trial. The size depends on the window used: 0.79% to 0.53% counting the first guideline change, 0.79% to 0.45% counting the follow-up too. [17][16]
05

Whether treating skin inflammation early prevents food allergy in general.

The recent randomised evidence is strongest for egg. Peanut, cow's milk and tree nuts are not settled, and cow's milk remains inconclusive. [9][8]
06

Why early peanut does so much less in Japan.

Peanut is eaten less there and allergy is rarer, but the review cannot separate diet, genetics and how the trials were run. [8]
07

What decides which babies get colonised by the harmful staph strains.

Early colonisation by strains carrying a working toxin-signalling system strongly predicts later eczema, and nothing yet explains who gets them. [29]
08

Whether topical steroid withdrawal is a distinct condition.

It has no official recognition and no agreed diagnostic criteria, while a large online community describes it in detail and rates it far more confidently than the evidence allows. [44]

04 Hope carry this

Parents in Britain and the US spent years being told to keep peanut away from babies, and peanut allergy in US children rose from 0.4% to more than 2%. The advice was reversed, and across 48 clinics peanut allergy in the under-threes fell from 0.79% to 0.45%.