- Design
- Nationwide population-based cohort, 5-year exposure and 6-year outcome periods
- Population
- 685,389 South Korean adults without diabetes at baseline
- Primary outcome
- Incident type 2 diabetes
- Effect
- Any use: no association; potent aHR 1.15 (1.04–1.26); ≥6 months aHR 1.45 (1.25–1.67)
This South Korean national insurance cohort followed 685,389 adults without diabetes. Topical corticosteroid use was measured over five years, by potency, cumulative duration and number of prescriptions, and new type 2 diabetes was recorded over the following six years.
Any topical steroid use was not associated with diabetes overall. The risk rose with exposure: potent steroids had an adjusted HR of 1.15 (95% CI 1.04–1.26), ten or more prescriptions 1.26 (1.12–1.42), and six months or more of cumulative use 1.45 (1.25–1.67). Low-potency use, shorter duration and fewer prescriptions were associated with a slightly lower risk.
Topical steroids are usually thought of as metabolically safe, and for short courses of mild preparations these data agree. The signal sits in potent and prolonged use, and the dose–response pattern makes it harder to dismiss. That matters in India, where potent steroid combination creams are widely bought over the counter and used for months. The study cannot separate the effect of the steroid from the disease being treated, but checking glucose in long-term potent users is cheap.
- Ask every patient with chronic dermatosis how long and how much potent steroid they have used, including over-the-counter creams
- Check HbA1c or fasting glucose in patients with six months or more of potent topical steroid use
- Use steroid-sparing agents such as topical calcineurin inhibitors for long-term maintenance where suitable
- Limit potent steroids to defined courses with a planned step-down
- Do not withhold short courses of appropriate-potency steroids on these grounds
Why it matters
Topical steroids are treated as metabolically neutral, and prolonged potent use may not be.
Don't overread it
This was observational — it shows an association with heavy use, not that topical steroids cause diabetes.
The statistics, in plain English
An adjusted hazard ratio of 1.45 means the rate of new diabetes was about 45% higher with six months or more of use, after accounting for measured factors. Because diabetes is common, even a modest relative increase adds up across many patients. The lower risk seen with low-potency use is likely confounding by who gets prescribed what, not a protective effect.
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