The results prompt new questions about how contraception may affect skin health.
Women using hormonal intrauterine devices face a significantly higher risk of developing rosacea compared to those using nonhormonal (copper) IUDs.
Researchers conducted a multicentre cohort study using data from the TriNetX network, examining women aged 18 to 50 years who received IUDs and were later diagnosed with rosacea. The results were published as a brief report in the Journal of the American Academy of Dermatology.
Researchers tracked rosacea diagnoses at one, three, and five years after insertion and compared rates between hormonal and nonhormonal IUD users. Women who had rosacea prior to insertion or had their IUDs removed during the study period were excluded.
Only one of the two hormonal IUDs available in Australia (Mirena) was part of the study. All three IUDs looked at in the study were levonorgestrel-releasing intrauterine systems – known as Liletta, Mirena and Skyla.
At one year, women using Liletta had the highest rate of rosacea, with 634 cases per 100,000 person-years and a 67% higher risk compared to copper IUD users. Mirena followed with a 39% increased risk. Skyla users saw a smaller, but non-significant, increase.
These patterns continued at across the follow-up period. After five years, Liletta users still showed the highest rosacea incidence, nearly 79% higher than copper IUD users. Mirena users also maintained a significantly increased risk of 47%. Skyla remained statistically comparable to the non-hormonal option throughout the study period.
The results support earlier concerns about how hormonal contraceptives interact with the body’s androgen and estrogen receptors, potentially triggering chronic inflammatory skin conditions. The study highlights the importance of considering dermatologic effects when selecting contraceptive methods.
“In contrast, nonhormonal IUDs exhibited lower [incidence rates], indicating that nonhormonal options may less likely influence rosacea development,” the authors wrote.
“Skyla did not show a significant difference from the copper IUD, suggesting a slower rate of levonorgestrel could play a role.”
Journal of the American Academy of Dermatology, February 2025
