Rosacea: Independent Evidence on Treatments, Redness & Funding

Table of contents

• Evidence summary
• What rosacea is, and how it is classified
• How the treatments are thought to work
• What works and what does not
• Benefits by claim
• Risks and all side effects
• Triggers, interactions and self-care
• Who should take extra care
• Doses studied and label doses
• Follow the money: who makes it and who funded the evidence
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

Independence tiers: Tier 1 (independent) = no identifiable financial stake in the result, such as public bodies and unfunded reviews; Tier 2 (indirect ties) = pooled evidence that relies on company-funded trials, or distant ties; Tier 3 (interested party) = a society, patient group or device-linked stake, not the direct seller; Tier 4 (self-interested) = produced, funded, sponsored or authored by the maker. Credibility runs from A (highest) to D (treat as advertising).

Convergence and clustering. The strongest signal is that the 2024 independent network meta-analysis, the Cochrane review and the NHS guidance all treat topical ivermectin, azelaic acid and metronidazole as first-line options for bumps and pustules. They disagree on the order, so no single winner is shown. Six of the twelve sources are maker-funded or maker-run, and a seventh, the National Rosacea Society, carries a Galderma-sponsored post. Our verdict therefore reflects a UK and US view, shaped by one company's products.

What this means: The independent evidence supports a real benefit for several topical creams on bumps and pustules, and for brimonidine on redness. The company-run trials that built the market are the weakest part of the record, and they make up most of the evidence for brimonidine and the 40 mg doxycycline dose. Where independent reviews reach the same conclusion, that conclusion is the safest one to rely on.

What rosacea is, and how it is classified

The NHS describes rosacea as a long-term skin condition that mainly affects the face. It is more common in women and in people with lighter skin, though symptoms can be worse in men. Early signs include flushing or warmth across the nose, cheeks, forehead, chin, neck and chest. Later signs can include constant redness, visible broken blood vessels, small pink or red bumps and, over many years, thickened skin mainly on the nose (NHS).

From subtypes to phenotypes

The older approach sorted patients into subtypes. The 2017 ROSCO panel recommended moving away from subtypes and classifying rosacea by the features a patient actually has. The panel judged two features diagnostic on their own: persistent centrofacial redness with periodic flares, which 17 of 17 panellists agreed on, and phymatous (thickened) skin changes, agreed by 13 of 17. Flushing, visible blood vessels and bumps were not diagnostic alone (Tan et al. 2017).

The 2019 update, by a panel of 19 dermatologists and 2 ophthalmologists, used at least 75% agreement. It grouped features into diagnostic (phymatous changes and persistent erythema), major (flushing or transient erythema, papules and pustules, telangiectasia) and minor (burning, stinging, dry sensation, oedema), plus ocular features such as lid-margin telangiectasia, blepharitis, keratitis, conjunctivitis and anterior uveitis (Schaller et al. 2019/2020). The 2017 panel's grant record lists Galderma, which sells rosacea products; the 2019 update's funding statement was not seen. Clinicians can use the classification, but the grading of which features matter most is a sponsor-funded opinion, not a neutral test.

The NHS page does not describe subtypes, and it lists look-alike conditions that should be considered, including acne, contact and seborrhoeic dermatitis, lupus and keratosis pilaris (NHS). A confident diagnosis, especially where eyes or thickened skin are involved, is a job for a GP or dermatologist.

How the treatments are thought to work

The cause of rosacea is not known. The NHS states it is not caused by poor hygiene and is not contagious (NHS). The leading treatments act on inflammation, on blood-vessel tone, or on the skin's microbes, and the mechanisms are only partly proven.

• Ivermectin 1% cream is licensed only for inflammatory lesions (papules and pustules), not for redness (Soolantra label). We did not verify a mechanism from a source we opened.
• Brimonidine gel is licensed for persistent redness, and the trials measured its effect over hours after each application (Mirvaso label; Fowler 2013). The reviews we opened do not describe its mechanism in detail.
• Doxycycline 40 mg is licensed as an anti-inflammatory dose. The label says plasma levels are below those needed to treat bacterial disease, and it warns that resistance may develop if the drug is misused (Oracea label).
• Lasers and IPL reduce visible redness and vessels in some patients in small trials. The 2025 review reports these effects but does not test a mechanism in detail (Nguyen 2025).

What works and what does not

Benefits by claim

Does ivermectin cream clear rosacea bumps?

Partly. Cochrane found topical ivermectin better than placebo in two studies (Cochrane 2015). In the 2016 network meta-analysis, which Galderma sponsored, ivermectin 1% beat metronidazole 0.75% with a risk ratio for treatment success of 1.17 (95% credible interval 1.08 to 1.29), and azelaic acid 15% twice daily with a risk ratio of 1.25 (1.14 to 1.37) (Siddiqui et al. 2016). One of its authors, Linda Stein Gold, is listed as a Galderma investigator. The later, unfunded 2024 network meta-analysis found ivermectin comparable to azelaic acid 15% and metronidazole, not clearly superior (Shaheen et al. 2024).

Does metronidazole work for bumps and pustules?

Yes, with moderate-quality evidence. Cochrane found topical metronidazole better than placebo for papulopustular rosacea, from three trials (Cochrane 2015). The 2024 network meta-analysis placed it in the same group as ivermectin and azelaic acid 15% (Shaheen et al. 2024). We did not check the funding of each metronidazole trial, so this rating is based on the review's own grading.

Does azelaic acid work?

Cochrane rates the benefit as high-quality, from four trials versus placebo (Cochrane 2015). The 2024 network meta-analysis gave azelaic acid 20% the highest ranking, with an odds ratio of 8.54 (95% CI 2.48 to 29.45) against placebo. The authors note the wide interval. The same review found azelaic acid 15% was the only topical agent with a significant increase in adverse events, odds ratio 1.95 (95% CI 1.30 to 2.93) (Shaheen et al. 2024).

Does brimonidine reduce redness?

Yes, for hours at a time. Cochrane rated brimonidine's effect on erythema as high quality, at three hours (Cochrane 2015). The two pivotal trials of 0.5% brimonidine tartrate gel, run by Galderma-linked authors, reported a benefit over vehicle; we could not verify the timing, day schedule or P values from the full report (Fowler et al. 2013). The FDA label describes the product for persistent, non-transient redness in adults 18 and over (Mirvaso label).

Does oxymetazoline (Rhofade) reduce redness?

The FDA approved oxymetazoline 1% cream in 2017 for persistent facial redness associated with rosacea (Aclaris press release, 2018). The company that acquired the rights, Aclaris Therapeutics, announced the deal closed on 30 November 2018, with the price not stated. We could not open the pivotal trial reports, so we cannot say how they were funded or how large the effect was. Treat the efficacy claim as unverified by independent work.

Does oral doxycycline help?

Yes, for bumps and pustules. Cochrane rated 100 mg doxycycline as better than placebo, with high-quality evidence. It found 40 mg about as effective as 100 mg, with fewer side effects; we could not verify the review's grade for that comparison (Cochrane 2015). The largest 40 mg study, ORCA, was sponsored by Galderma Laboratories. Of 1,421 patients in the intent-to-treat group, about 75% of the 826 monotherapy patients who completed the study were clear or near clear by week 12 (Galderma, 2010). The study had no comparator group, so it cannot show benefit over placebo.

Do lasers and IPL help redness and visible blood vessels?

There is some evidence, but it is weak. A 2024 meta-analysis of four studies with 141 people found IPL and pulsed-dye laser similar for erythema, and found a significantly higher share of patients with more than 75% clearance on IPL (Zhai et al. 2024). We could not see this study's funding or conflict statements. The sign convention for the clearance result is unclear in the paper, so we report the direction the authors state. The 2025 network meta-analysis of 25 RCTs found PDL with oxymetazoline was the top-ranked treatment for telangiectasia, with a mean difference of −0.58 (95% CI −1.03 to −0.14), while radiofrequency microneedling beat PDL on satisfaction and erythema (Nguyen et al. 2025). That review was supported by a German laser society and two authors had received lecture fees from a device maker.

Risks and all side effects

The most common side effects in the trials and labels are local. For ivermectin 1% cream, we did not check the label's side-effect rates line by line (Soolantra label). For brimonidine gel, the most common reactions on the US label are erythema, flushing and skin burning sensation (Mirvaso label). For oral doxycycline 40 mg, we did not check the label's reaction rates line by line (Oracea label).

Azelaic acid 15% was the only topical agent with a significantly higher rate of adverse events than placebo in the 2024 network meta-analysis (odds ratio 1.95, 95% CI 1.30 to 2.93), and the authors note that some agents had limited safety data (Shaheen et al. 2024).

Triggers, interactions and self-care

The NHS lists alcohol, spicy food, hot drinks, sunlight, hot or cold temperatures, aerobic exercise such as running and stress as common triggers. It advises daily SPF 30 sunscreen, gentle skincare, and avoiding heat and humidity where possible (NHS). The NHS does not rank how strongly each trigger drives flares, and we did not find a graded trigger study in our search, so we rate trigger avoidance as insufficiently tested.

We did not find independent interaction studies for these products in this search, and we did not verify interaction sections from the labels. Check any new medicine with a pharmacist, especially antibiotics and any medicine that makes skin more sensitive to sunlight. Because the NHS lists alcohol and spicy food as triggers, people who notice flares after drinking may want to note this and raise it with their clinician.

Who should take extra care

• Anyone with eye symptoms should be assessed urgently, as above (NHS).
• People taking oral antibiotics or with recurrent infections should discuss doxycycline with their clinician, because the Oracea label says the product is not for infections and warns of resistance (Oracea label).
• People with unexplained facial redness or thickened skin should see a dermatologist, because look-alike conditions include lupus and dermatitis (NHS).
• Pregnancy, breastfeeding and children: we did not verify label guidance on these groups for each product in this search. Ask a pharmacist or clinician before starting.

Doses studied and label doses

The doses below are the ones used in trials or stated on US labels. They are not personal instructions. Follow the prescriber's directions for any real treatment.

• Ivermectin 1% cream: label says once daily on affected areas of the face. The pivotal trials used 12 weeks of treatment, and some patients were treated for about a year (Soolantra label; Stein Gold 2014).
• Metronidazole 0.75% cream: twice daily in the 2016 and 2024 comparisons (Siddiqui 2016).
• Azelaic acid 15% gel: twice daily in the 2016 comparisons (Siddiqui 2016).
• Brimonidine 0.5% tartrate gel: once daily for 4 weeks, followed by 4 weeks of follow-up in the pivotal trials (Fowler 2013). The marketed product is a 0.33% gel, and the label gives its own directions (Mirvaso label).
• Doxycycline 40 mg modified-release (Oracea): one 40 mg capsule once daily in the morning, on an empty stomach, at least 1 hour before or 2 hours after meals (Oracea label).

Product quality. We did not find independent product-testing results for rosacea creams in this search. The Oracea label names Catalent Pharma Solutions as manufacturer; the country of that contractor was not verified.

Follow the money: who makes it and who funded the evidence

Who makes and sells the main products

• Galderma Laboratories, L.P. is the labeller for Soolantra (ivermectin 1% cream), Oracea (doxycycline 40 mg) and Mirvaso (brimonidine gel), and is based in Dallas, Texas, according to the DailyMed labels (Soolantra label; Oracea label; Mirvaso label).
• Galderma Group AG is the listed parent, traded on SIX. The company listed in 2024 (EQT, March 2026).
• Ownership. EQT, the private equity owner, fully exited Galderma Group AG on 13 March 2026, selling about 34.0 million shares for about CHF 4.9 billion in gross proceeds. An EQT fund affiliate received about CHF 1.3 billion. L'Oréal holds a 20% stake, built through two tranches after the listing (EQT, March 2026).
• Oxymetazoline (Rhofade). Allergan sold the worldwide rights to Aclaris Therapeutics, which closed on 30 November 2018. The price was not stated (Aclaris press release).

Who funded the trials and guidance

• Siddiqui et al. 2016: sponsored by Galderma, with research grants to two authors and Galderma investigator status for a third (Siddiqui 2016).
• ORCA doxycycline study: sponsored by Galderma Laboratories L.P. (Galderma 2010).
• ROSCO 2017 and 2019 panels: the 2017 grant record lists Galderma; honoraria not verified; the 2019 update's funding statement was not seen (Tan 2017; Schaller 2019/2020).
• Brimonidine pivotal trials: the report was on behalf of the Brimonidine Phase III Study Group, with two Galderma R&D authors (Fowler 2013). A later safety analysis of the same trials, seen only in search results and not opened, is reported to disclose Galderma employees and Galderma support for the manuscript.
• Ivermectin pivotal trials: published on behalf of the Ivermectin Phase III Study Group, with Galderma R&D affiliation. The visible page did not show a funding statement (Stein Gold 2014).
• Cochrane 2015: 106 included studies and 13,631 participants; we did not verify the review's funding breakdown (Cochrane 2015).
• Shaheen et al. 2024: no conflicts disclosed; funding statement not seen (Shaheen 2024).
• Laser review 2025: supported by the German Society for Dermatological Laser Medicine; two authors had lecture fees from Cynosure Lutronic (Nguyen 2025).
• National Rosacea Society: sustained by donations from individuals and industry. Industry money funds awareness and education, and the Society says it alone is responsible for those materials (NRS funding page). A May 2017 post states it was sponsored by Galderma Laboratories, L.P. (NRS 2017 post).

Documented integrity events

We did not find documented regulatory bans, warning letters, FTC actions or recalls for the rosacea products listed above in our search. We did not check the FDA warning-letter database or court records, so this is a gap rather than a finding of none.

Related research

Other skin-and-inflammation evidence reviewed by Pure City Research:

• Eczema prevention and management: evidence and safety guide
• Eczema supplements: evidence, uncertainty and safety
• Astaxanthin: skin, UV response, exercise and safety evidence
• Collagen: skin, joint and bone evidence, risks and funding
• Glutathione: skin trials, antioxidant markers and injection safety
• Melanoma: skin and nail symptoms, biopsy, staging and treatment

Frequently asked questions

Sources and funding notes

What we could not verify: the Cochrane risk ratios and funding breakdown (publisher full text blocked); the Shaheen and Zhai funding statements; the Tan honoraria and voting statements and the 2019 ROSCO funding; the adverse-event rates on the Soolantra, Mirvaso and Oracea labels; The funding statement for the 2014 ivermectin pivotal publication (Stein Gold et al.); the funding statements for the 2013 brimonidine trial report and the oxymetazoline pivotal trials; the Galderma-sponsored ORCA trial report itself (we used the company's press release); the Cochrane review's own funding and declarations of interest; the 2019 ROSCO conflict statement in its appendix; the British Association of Dermatologists 2021 guideline, which we could not open; the NICE brimonidine assessment (the page returned an error); the Oracea EMA referral (the PDF could not be read); the authors' countries for the 2016 ivermectin review, the 2024 azelaic-acid network meta-analysis and the 2024 IPL meta-analysis; and the location of Catalent Pharma Solutions. Two PubMed records (the 2017 and 2019 ROSCO papers) were rate-limited during this review, so we cite their DOI pages instead. The 2024 network meta-analysis reports 8,208 participants in its abstract and 8,075 in its results, so we avoid quoting a total. We found no independent trial of the creams on eye disease.

1. NHS, Rosacea — UK public health guidance; funded by UK public funds.
2. Van Zuuren EJ et al., Cochrane Interventions for rosacea, 2015 — systematic review of 106 RCTs (13,631 participants); funding breakdown not verified.
3. Cochrane featured review, June 2015 — summary of the same review; no funding statement shown.
4. Shaheen et al., network meta-analysis, JAAD International 2024 — 19 RCTs; no conflicts disclosed; funding statement not seen.
5. Siddiqui K, Stein Gold L, Gill J, SpringerPlus 2016 — 19 RCTs; sponsored by Galderma; grants to two authors.
6. Stein Gold L et al., J Drugs Dermatol 2014 — two pivotal ivermectin trials; funding statement not visible; Galderma R&D affiliation.
7. Fowler J et al., J Drugs Dermatol 2013 — two pivotal brimonidine trials; Galderma R&D affiliation on two authors.
8. Galderma, ORCA results press release, 2010 — maker's own release; sponsored by Galderma Laboratories L.P.
9. Tan J et al., ROSCO panel, Br J Dermatol 2017 — grant record lists Galderma; honoraria not verified.
10. Schaller M et al., ROSCO 2019 panel, Br J Dermatol (vol 182, 2020) — funded by Galderma; conflict statement in appendix not seen.
11. Zhai Q et al., J Cosmet Dermatol 2024 — 4 studies, 141 participants; funding and conflict statements not seen.
12. Nguyen L et al., J Dtsch Dermatol Ges 2025 — 25 RCTs; supported by the German Society for Dermatological Laser Medicine; lecture fees from Cynosure Lutronic for two authors.
13. National Rosacea Society, How the NRS is funded — patient group; funded by individual and industry donations.
14. National Rosacea Society, sponsored post, May 2017 — states sponsorship by Galderma Laboratories, L.P.
15. Soolantra (ivermectin) cream label, DailyMed — labeller Galderma Laboratories, L.P., Dallas; maker-written, FDA-reviewed.
16. Mirvaso (brimonidine) gel label, DailyMed — labeller Galderma Laboratories, L.P.; rebound statement; adverse reaction rates.
17. Oracea (doxycycline 40 mg) label, DailyMed — labeller Galderma Laboratories, L.P.; dosing and antibacterial caveat.
18. Aclaris Therapeutics press release, December 2018 — acquisition of Rhofade rights from Allergan; closed 30 November 2018; price not stated.
19. EQT, Galderma Group AG full exit, 13 March 2026 — ownership change; L'Oréal 20% stake; SIX listing.

Educational research reviewed 9 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.