Most adults with atopic dermatitis get control on standard therapy. A minority do not, and the reflex when that happens is escalation — a stronger topical, then a systemic, then a biologic. The American Academy of Dermatology has issued a Good Practice Statement arguing that the first move should be different: reconsider whether the diagnosis is right, and whether something else is running alongside it.
An expert multidisciplinary workgroup applied GRADE methodology, reviewing indirect evidence, weighing benefits and harms of investigation, and reaching consensus on the diagnostic workup of adults with presumed atopic dermatitis unresponsive to optimised treatment. The document's own limitation section is unusually candid: direct empirical data on diagnostic workup strategies for treatment-resistant atopic dermatitis do not exist, and applicability varies with access to dermatology and allergy specialist care.
That honesty is worth respecting rather than discounting. This is a statement about clinical reasoning, and the reasoning is sound whether or not a trial has tested it. The conditions that masquerade as refractory eczema are well known and are individually treatable in ways that escalating immunosuppression is not: allergic contact dermatitis superimposed on true atopic dermatitis, cutaneous T-cell lymphoma, scabies, psoriasis with an eczematous morphology, drug eruption. Each of those gets worse, not better, when the answer to non-response is another systemic agent.
For Indian practice the differential shifts in emphasis and the argument gets stronger. Scabies is common and frequently treated for months as eczema. Chronic dermatophyte infection, often modified by over-the-counter topical steroid-antifungal combinations that are widely sold here, produces a picture that looks nothing like classical tinea and everything like refractory eczema. And the cost of escalating to a systemic agent is borne directly by the patient. A scraping and a careful drug and product history cost almost nothing and change the answer more often than the next drug does.
- Treat non-response to optimised therapy as a prompt to reassess the diagnosis, not to escalate.
- Consider allergic contact dermatitis superimposed on atopic dermatitis — patch testing is the answer.
- Rule out scabies and dermatophyte infection before starting a systemic agent.
- Keep cutaneous T-cell lymphoma in mind in an adult whose eczema started late and will not settle.
- Ask specifically about over-the-counter steroid-antifungal combination creams, which distort the clinical picture.
The statistics, in plain English
A Good Practice Statement is the weakest formal output GRADE allows, and it is used deliberately when a recommendation is obviously sensible but has never been and probably never will be tested — nobody will randomise patients to 'reconsider the diagnosis' versus 'do not'. The workgroup says the evidence is indirect and the basis is consensus. Read it as codified clinical judgement from people who see this problem often, which is a legitimate thing to act on, rather than as an evidence-based recommendation in the usual sense.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for dermatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free