A trigger log only works if you read it against the right delay. An irritant reaction to washing-up liquid shows within hours. A contact allergy is slow enough that clinics tape the test substances to your back for 48 hours before reading them. NICE's guideline on atopic eczema in children under 12 names five categories worth recording: irritants such as soaps and detergents, skin infections, contact allergens, food allergens and inhalant allergens (NICE CG57, 2025). Log those against a daily severity score for four weeks, then read each category over its own timescale.
Why single-exposure recall fails
Memory anchors to whatever was vivid and recent, which is why the new shower gel gets blamed and the fortnight of bad sleep does not. The harder problem is that triggers act on different clocks.
Irritants are a direct chemical insult to a barrier that is already leaking, and the response follows within hours, so recall usually gets those right. Contact allergy is different. It is a delayed, cell-mediated reaction, which is why patch testing leaves the substances on for 48 hours, checks again at a third appointment, and warns that "a reaction can occasionally occur after the 3rd appointment" (British Association of Dermatologists). If a clinic needs two visits days apart, one evening's recall will not find it.
Then there is the base rate. NICE describes atopic eczema as "typically an episodic disease of exacerbation (flares, which may occur as frequently as 2 or 3 per month) and remissions" (NICE CG57, 2025). Flares arrive on their own schedule. Match one flare to one exposure the day before and you have a story you cannot distinguish from the condition doing what it does anyway.
What to log each day, and how to score it
Use a severity score that already exists. The Patient-Oriented Eczema Measure asks seven questions, each scored 0 to 4, for a total of 0 to 28, banded from clear or almost clear at 0 to 2 through moderate at 8 to 16 (Charman et al, 2004; Centre of Evidence Based Dermatology). It scores how often symptoms occurred rather than how the skin looks, which is the judgement you can make about yourself consistently.
Around that weekly score, record five things daily at a fixed time. Itch, 0 to 10. Nights disturbed, counted as wakings rather than rated, because a count is harder to inflate. Affected areas ticked off a fixed list (hands, wrists, inner elbows, backs of knees, neck, face, trunk), so spread shows as well as intensity. Exposures, filed under NICE's five categories plus heat and stress, which the NHS names alongside soap, detergent, pets, fabrics and house dust mites. And one line on what the day cost you: NICE notes no necessary relationship between eczema severity and its impact on quality of life.
Add a photograph once a week: same room, same light, same distance, same time of day. Phone-camera skin tone moves more with the lighting than with the inflammation.
How long to log before you read it
Four weeks is the floor, and the reason is arithmetic rather than dermatology. At two or three flares a month, a fortnight might hold one flare and one settled stretch. One of each cannot separate a trigger from the disease's own rhythm. Four weeks covers several cycles, so a pattern can repeat rather than simply occur. POEM was shown to be sensitive to change across a four-week in-clinic period (Charman et al, 2004), a window over which a validated eczema score has been shown to register change.
A stress and sleep log can be read at two weeks, because that lag is measured in hours. Either way, the log runs for as long as the slowest thing in it takes to show.
Reading the log without fooling yourself
Start with your baseline range. Take the first two weeks, find your median daily itch score, and note the highest and lowest ordinary days. Anything inside that range is noise you already produce, and without it you will read normal variation as a response to whatever you did that morning.
Then test candidates in both directions. A real suspect appears before most of your worst days and is absent before most of your best. Checking only the bad days finds the things you do daily: you sleep in your bedroom before every flare and before every good week, so dust exposure passes a one-directional test whatever it is doing.
What survives is a hypothesis, not a trigger. Confirming one means removing all your candidates completely and at once, then reintroducing them one at a time every three to four days, the elimination and reintroduction framework. Keep scoring through both phases with the same daily log fields. A confirmed trigger reproduces: it should fail twice in the same direction, at the interval that trigger works on. Once is coincidence with better paperwork.
What the evidence says about the common classes
Irritants are NICE's first category and the cheapest to act on, which is why a room-by-room audit comes before any dietary experiment. Hard water belongs in that category: it reacts with soap to leave surfactant deposits that rinsing does not remove, though an ion-exchange softener added to usual care gave no severity benefit in a trial of 336 children (Thomas et al, 2011).
House dust mite is where logging and treating come apart. A Cochrane review of mite reduction and avoidance pooled seven trials and 324 participants and found the evidence "falls far short of a basis for influencing clinical practice" (Nankervis et al, 2015). Mite exposure may still be your trigger. What is unproven is that covers and vacuuming treat established eczema, so a null result there does not clear the allergen.
Food is where the guidance is more cautious than the internet. NICE's diagnostic flag is immediate symptoms after eating a particular food, or moderate to severe eczema in babies and young children that "has not been controlled by optimum management". It also puts a clock on exclusion: refer for specialist dietary advice if a cow's-milk-free diet has run beyond eight weeks.
Keeping it sustainable
The failure mode is abandonment in week two. Attach the entry to something that already happens at a fixed time, and put the weekly items on one named day. If you miss a day, leave it blank. Filling it in from memory reintroduces the error the log exists to remove.
References
- National Institute for Health and Care Excellence. Atopic eczema in under 12s: diagnosis and management. NICE clinical guideline CG57, 2025. https://www.ncbi.nlm.nih.gov/books/NBK571009/
- NHS. Atopic eczema: causes. https://www.nhs.uk/conditions/atopic-eczema/causes/
- Charman CR, Venn AJ, Williams HC. The patient-oriented eczema measure. Archives of Dermatology, 2004;140(12):1513-1519. https://doi.org/10.1001/archderm.140.12.1513
- Centre of Evidence Based Dermatology, University of Nottingham. Patient Oriented Eczema Measure. https://www.nottingham.ac.uk/research/groups/cebd/resources/poem.aspx
- Nankervis H, et al. House dust mite reduction and avoidance measures for treating eczema. Cochrane Database of Systematic Reviews, 2015. https://pmc.ncbi.nlm.nih.gov/articles/PMC8407038/
- Thomas KS, et al. A randomised controlled trial of ion-exchange water softeners for the treatment of eczema in children. PLoS Medicine, 2011. https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1000395
- British Association of Dermatologists. Patch testing. https://www.bad.org.uk/pils/patch-testing/