Pick two things you are using — or thinking of using — and get what the evidence actually says about putting them on the same face. Not a rule someone invented for a blog post. The study, the drug label or the guideline statement, named and linked, or an honest admission that nobody has looked.
Here is what we found while building this, and it is the opposite of what most “don’t mix these” lists say. Of the 24 combinations we researched properly, only three justify genuinely keeping the two apart, and two more are questions for a prescriber. Everything else is either actively beneficial, harmless, or simply unstudied — and unstudied gets said out loud here rather than dressed up as a warning.
This is a research reference, not medical advice.
Everything here is a plain-English summary of what published guidelines and studies say. It is not a diagnosis, not a treatment plan, and not a substitute for a conversation with a doctor, dermatologist, pharmacist or other qualified clinician who can actually examine your skin and knows your medical history.
Nobody here is a clinician. This site is researched and written by a non-expert who reads the primary sources carefully and links every claim to them so you can check the work yourself. That is a real method, and it is also a real limit. Where a guideline says something, we quote the guideline. Where the evidence is thin or missing, we say that instead of filling the gap with a guess.
Prescription medicines, in-clinic procedures and anything you swallow are decisions for you and a clinician together. Do not start, stop, change the dose of, or combine any treatment on the strength of a web page — including this one. If something on your skin is painful, spreading, scarring or distressing you, please see someone in person.
Every pairing we have researched properly
These 26 combinations have named evidence behind them — a trial, a stability study, a drug label or a guideline statement. Pick anything else in the tool above and it will tell you honestly that nobody has studied it.
Fine, with a caveat Benzoyl peroxide + Tretinoin (prescription retinoid)
In short
Real chemistry, and it depends entirely on which tretinoin you have.
What the evidence says
This is the one 'don't mix' rule with a genuine laboratory finding behind it. Martin and colleagues mixed conventional tretinoin gel with benzoyl peroxide and exposed it to light: more than 50% of the tretinoin degraded in about two hours and 95% within 24. But that was 1998, tested on a plain gel, and formulation has moved on. Tretinoin gel microsphere retained 87–99% of its tretinoin when mixed with a benzoyl peroxide product, and an optimised aqueous tretinoin gel retained 100% at seven hours. The decisive evidence is regulatory: the FDA has approved a single cream containing both, which works by micro-encapsulating each active in separate silica shells.
Commonly repeated, and not quite right
Commonly repeated: that Martin's 1998 study showed benzoyl peroxide degrades tretinoin, adapalene and tazarotene. What the paper actually tested: adapalene and tretinoin only. Tazarotene was not in it. Tazarotene photostability comes from separate work by Hecker and colleagues.
In practice
With plain generic tretinoin cream or gel, use benzoyl peroxide in the morning and tretinoin at night — it costs nothing and removes the question. With microsphere, optimised-vehicle or encapsulated products, there is no stability reason to separate them.
Sources
Fine together Adapalene (Differin) + Benzoyl peroxide
In short
Fine together — there are FDA-approved products containing both.
What the evidence says
The same 1998 experiment that found tretinoin degrading found adapalene showed, in the authors' words, 'remarkable stability' with benzoyl peroxide, with and without light. Two fixed-dose prescription products combine them in one tube, which settles the stability question as far as it can be settled.
In practice
Layer them, or use a combination product. The thing to watch is irritation, not chemistry — this pairing is more drying than either alone.
Sources
Fine together Topical antibiotic (clindamycin, erythromycin) + Benzoyl peroxide
In short
Not just fine — this is the pairing guidelines actively want you to use.
What the evidence says
This is the strongest evidence on this page, and it runs the opposite way to the usual 'don't mix' instinct. Benzoyl peroxide has no documented resistance in the bacteria involved in acne, so it suppresses the resistant population that antibiotic monotherapy selects for. In one clinical comparison, clindamycin-resistant bacterial counts rose to over 1600% of baseline on clindamycin alone, but stayed at or below baseline when benzoyl peroxide was added. The AAD guideline makes fixed-dose antibiotic-plus-benzoyl-peroxide its own strong recommendation, and separately notes that topical antibiotic monotherapy is not recommended.
In practice
Use them together, always. If your prescription is an antibiotic alone, that is worth asking your prescriber about.
Sources
Fine together Benzoyl peroxide + Niacinamide
In short
Fine together, and a randomised trial suggests the combination is better.
What the evidence says
A 12-week split-face randomised trial compared 2.5% benzoyl peroxide plus 5% niacinamide against 2.5% benzoyl peroxide alone. The combination reduced non-inflammatory lesions by 51% versus 34%, with no difference in tolerability and no sign of chemical trouble.
Commonly repeated, and not quite right
Commonly repeated: that niacinamide converts to nicotinic acid and causes flushing when combined with acids or oxidisers. Where that comes from: hydrolysis rate studies from the early 1960s conducted at elevated temperature and extreme pH — conditions nothing like a face. A review of topical 10% nicotinamide reported no stinging or flushing in human subjects. The direct trial above found benefit, not a problem.
In practice
Layer them in either order. No need to separate by time of day.
Sources
Fine together Niacinamide + Vitamin C (L-ascorbic acid)
In short
Fine together — the incompatibility comes from 1960s solution chemistry, not from skin.
What the evidence says
The claim traces to Guttman's 1963 work describing a complex between nicotinamide and ascorbic acid in aqueous solution that accelerates degradation and yellows the mixture, and to a later kinetic study, also in solution. Neither models a modern low-water cosmetic vehicle and neither studied skin. Meanwhile products deliberately combining the two are in clinical use: a randomised double-blind trial of a gel-cream containing 10% nicotinamide with 5% magnesium ascorbyl phosphate ran against hydroquinone for melasma with no instability reported and no serious side effects.
Commonly repeated, and not quite right
Commonly repeated: that vitamin C and niacinamide 'cancel each other out' and must be used twelve hours apart. What the underlying papers actually studied: the two compounds dissolved together in water, under accelerated conditions, sixty years ago. No study has shown the effect on skin, and formulations combining them are used in trials.
In practice
Use them in the same routine if you like. If a vitamin C serum turns yellow-brown, that is the ascorbic acid oxidising on its own — which it does regardless of what else is on your face.
Sources
Nobody has tested this Benzoyl peroxide + Vitamin C (L-ascorbic acid)
In short
Nobody has tested this. The chemistry makes separating them a free, sensible hedge.
What the evidence says
We could not find a single stability study or clinical trial of ascorbic acid co-applied with benzoyl peroxide. The concern is a first-principles one: benzoyl peroxide generates radicals and L-ascorbic acid is a reducing agent, so mixing them in one film should oxidise the ascorbate. That is a reasonable inference. It is not a measurement, and neither ingredient becomes unsafe. Notably, benzoyl peroxide's documented topical interactions list does not include ascorbic acid.
In practice
Vitamin C in the morning, benzoyl peroxide in the evening. It costs nothing and sidesteps a question nobody has answered.
Sources
Keep them apart Benzoyl peroxide + Hydroquinone
In short
Separate them — this is one of the few 'don't mix' rules with real support.
What the evidence says
Hydroquinone oxidises readily; benzoyl peroxide is a strong oxidiser. The result is reported transient staining of the skin. Clinical references are direct about it: DermNet's hydroquinone guidance says not to use other topical medication, benzoyl peroxide particularly. A closely related and better-documented case is benzoyl peroxide with topical dapsone or sulfacetamide, which produces an orange-brown discoloration of skin and facial hair.
In practice
Different times of day, or different days. Be aware we found no trial or case series quantifying this specific staining — it is a well-recognised clinical caution with a clear mechanism rather than a measured effect, and the discoloration is described as temporary.
Sources
Fine, with a caveat Salicylic acid (BHA) + Tretinoin (prescription retinoid)
In short
No chemical problem. This is purely a question of how much irritation your skin will take.
What the evidence says
Nothing degrades. The AAD guideline's first good practice statement actively recommends combining topical therapies with different mechanisms of action, and gives salicylic acid a conditional recommendation alongside a strong one for retinoids. There is even trial evidence for a single gel containing tretinoin, clindamycin and glycolic acid — though notably, that study existed to test a way of reducing the irritation such a combination causes.
Commonly repeated, and not quite right
Commonly repeated: that acids 'deactivate' retinoids by lowering the pH. What is actually going on: tretinoin is not pH-inactivated by a separate product applied to skin, and combination products containing both a retinoid and an acid exist and have been trialled. The real and well-documented issue is cumulative irritation.
In practice
Stagger them — different nights, or acid in the morning and retinoid at night — and build up. If your skin stings, flakes or goes red, that is dose, not incompatibility.
Sources
Fine, with a caveat Adapalene (Differin) + Salicylic acid (BHA)
In short
No chemical problem. It is a question of cumulative irritation.
What the evidence says
Adapalene is the most chemically stable of the retinoids and there is no interaction to worry about. The guideline recommends combining topicals with different mechanisms of action. What it cannot tell you is how much your particular skin will tolerate.
In practice
Alternate nights, or acid in the morning and adapalene at night. Add one thing at a time so you know what caused what.
Sources
Fine, with a caveat Niacinamide + Tretinoin (prescription retinoid)
In short
Reasonable, but the specific claim is thinner than it sounds.
What the evidence says
We could not find a trial of niacinamide added specifically to a retinoid. What exists is adjacent: a split-face double-blind randomised trial found a moisturiser containing ceramides and niacinamide, used alongside topical acne treatment, beat a plain cream on lesion counts, hydration, sebum and skin pH, and concluded it decreased cutaneous irritation. Niacinamide is also known to support the skin barrier components that retinoids disrupt.
In practice
Fine to use together, and plausibly helpful. Just be aware the honest version of the claim is 'barrier-supporting moisturisers containing niacinamide improve tolerability of topical acne treatment', not 'niacinamide cancels retinoid irritation'.
Sources
Fine together Azelaic acid + Tretinoin (prescription retinoid)
In short
No interaction known, and no trial of the combination either.
What the evidence says
Azelaic acid is a simple saturated dicarboxylic acid with no redox or photolytic reactivity toward retinoids, so there is no mechanism for a problem. Cochrane compared them head to head and found little or no difference in treatment response (moderate certainty). Nobody has run a trial of the two together.
In practice
Layer them, or alternate. Azelaic acid is among the better-tolerated actives, so this is usually a comfortable pairing — and azelaic acid also works on the dark marks retinoids leave behind.
Sources
Fine, with a caveat Tretinoin (prescription retinoid) + Vitamin C (L-ascorbic acid)
In short
No evidence of harm. The pH argument is about one bottle, not two products.
What the evidence says
The usual objection is that L-ascorbic acid needs a pH below about 3.5 while retinoids are formulated near neutral. That is a formulation-stability argument about co-formulating them in a single product — it says little about applying two finished products minutes apart. A retinol and vitamin C combination has been evaluated histologically on skin, so co-formulation is at least attempted.
In practice
Vitamin C in the morning, retinoid at night is the conventional split and is perfectly reasonable. The real reason to separate is additive irritation, especially with L-ascorbic acid at 15–20%.
Sources
Fine together Azelaic acid + Niacinamide
In short
No evidence of any interaction — and no trial of the pair either.
What the evidence says
Neither the Cochrane review nor the large azelaic acid systematic review evaluated this combination. Both ingredients are chemically unreactive toward each other and both are well tolerated. Compounding pharmacies dispense gels containing both, which tells you no pharmacist regards them as incompatible — that is practice, not evidence, and we would rather say so.
In practice
Use them together in either order.
Sources
Nobody has tested this Azelaic acid + Benzoyl peroxide
In short
No combination trial exists. There is no mechanistic red flag.
What the evidence says
All the available data are head-to-head rather than combined. Cochrane found azelaic acid probably leads to a worse treatment response than benzoyl peroxide (moderate certainty). Azelaic acid is not obviously oxidisable by benzoyl peroxide, so there is no chemistry reason to avoid the pairing — but nobody has tested it.
In practice
If you use both, watch for irritation and introduce them one at a time. There is no evidence to promise you a benefit from combining them.
Sources
Fine together Benzoyl peroxide + Salicylic acid (BHA)
In short
Fine together, with one regulatory nuance worth understanding.
What the evidence says
Two randomised trials support co-use. Adding salicylic acid to a clindamycin-plus-benzoyl-peroxide regimen was tested directly. A later randomised double-blind study of a salicylic-acid-and-niacinamide product used alongside benzoyl peroxide found the group using both needed significantly less benzoyl peroxide for equivalent results.
Commonly repeated, and not quite right
Sometimes cited as a warning: that the US OTC monograph does not allow benzoyl peroxide and salicylic acid in the same product, therefore they should not be used together. What that rule actually is: a restriction on marketing a single combined over-the-counter product without new data. It permits only resorcinol-with-sulfur as a combination. It is a regulatory and data matter, not a finding that the two are incompatible on skin.
In practice
Use them. Watch for dryness — both are drying, and the combination more so.
Sources
Fine, with a caveat Glycolic or lactic acid (AHA) + Salicylic acid (BHA)
In short
No chemical interaction at all. This is dose, not incompatibility.
What the evidence says
No study has looked at simultaneous home use of two acids. What exists is comparative peel data — Cochrane found no difference between glycolic acid peels and salicylic-mandelic peels (low certainty), and several head-to-head peel trials exist. Two acids simply lower pH together and deliver additive keratolysis. The risk is barrier disruption from over-exfoliating.
In practice
Pick one for most days. If you use both, use them on different days, and stop if skin feels tight, shiny-tender or stings on water.
Sources
Ask your prescriber first Hydroquinone + Tretinoin (prescription retinoid)
In short
An established combination — but the classic version is a prescription with a steroid in it.
What the evidence says
The Kligman formula dates to 1975 and combines hydroquinone, tretinoin and a corticosteroid; an FDA-approved product exists for melasma. Tretinoin alone also has randomised evidence for post-inflammatory hyperpigmentation specifically in Black patients. But a 2025 meta-analysis of the triple formula found no statistically significant advantage over alternative treatments on standard pigmentation scores, with high heterogeneity.
In practice
This is clinician territory, not a routine you assemble yourself — the steroid component makes it short-term and supervised, and prolonged hydroquinone carries a risk of exogenous ochronosis, a blue-black darkening that should prompt stopping.
Sources
Fine together Topical antibiotic (clindamycin, erythromycin) + Tretinoin (prescription retinoid)
In short
Fine together — approved products contain both in one tube.
What the evidence says
A prescription gel combining clindamycin 1.2% with tretinoin 0.025% is FDA-approved for acne from age 12, and the AAD guideline makes fixed-dose retinoid-with-antibiotic its own strong recommendation on moderate-certainty evidence.
In practice
Two label points worth carrying over: products containing clindamycin should not be combined with erythromycin-containing products, and because these combinations contain no benzoyl peroxide, the guideline recommends using benzoyl peroxide somewhere in the regimen anyway to limit resistance.
Sources
Keep them apart Oral isotretinoin + Waxing, dermabrasion or laser resurfacing
In short
A labelled caution, not folklore. Wait six months.
What the evidence says
The isotretinoin label is explicit: wax epilation and skin resurfacing procedures such as dermabrasion and laser should be avoided during treatment and for at least six months afterwards, because of the possibility of scarring.
In practice
Threading, shaving or trimming instead. Tell any clinic doing laser, peels, dermabrasion or microneedling that you have taken isotretinoin and when you stopped.
Sources
Ask your prescriber first Oral doxycycline + Oral isotretinoin
In short
A labelled interaction. This is a prescriber's call, not yours.
What the evidence says
Both labels carry the warning from both directions. The isotretinoin label states that concomitant treatment with tetracyclines should be avoided, citing the risk of raised pressure around the brain (pseudotumor cerebri); the doxycycline label says the same in reverse.
Commonly repeated, and not quite right
Commonly over-generalised: that 'retinoids and doxycycline don't mix'. The actual warning is narrower: it names oral isotretinoin and the tetracycline class. Topical tretinoin or adapalene with oral doxycycline is a standard, guideline-supported combination.
In practice
If you have been prescribed both, raise it with the prescriber before taking them — it may be an oversight, or there may be a plan. Note this warning is about oral isotretinoin. Oral doxycycline with a topical retinoid is standard combination therapy with no such interaction.
Sources
Fine together Combined oral contraceptive + Spironolactone
In short
Routinely prescribed together, and the monitoring folklore has been studied.
What the evidence says
The AAD guideline conditionally recommends both, and they are frequently combined — the contraceptive also covers the pregnancy precaution that comes with spironolactone. On the hyperkalaemia worry, research found potassium monitoring of low usefulness in healthy young women taking spironolactone for acne, and the guideline's own remark says monitoring is not needed in healthy patients.
In practice
This is prescription territory, so it is a conversation with your prescriber rather than a decision you make. Monitoring is still appropriate with kidney or heart disease, older age, or other medications that raise potassium — and some contraceptives are themselves mildly potassium-sparing.
Sources
Fine together Benzoyl peroxide + Moisturiser or sunscreen
In short
Use the moisturiser and the sunscreen. They do not blunt the treatment.
What the evidence says
In a 12-week open-label study, benzoyl peroxide 5% gel used with a cleanser and an SPF 30 moisturiser still reduced acne bacteria by 89% at week one, with most participants reporting less irritation and considering the moisturiser necessary. Every benzoyl peroxide product label reviewed tells users to apply broad-spectrum sunscreen when sun exposure cannot be avoided.
In practice
The genuinely true piece of benzoyl peroxide folklore is the laundry one: it bleaches hair and coloured fabric. White pillowcases and towels, and let it dry before you dress.
Sources
Nobody has tested this Benzoyl peroxide + Sulfur
In short
Barely studied, in either direction.
What the evidence says
Sulfur is the thinnest-evidenced active here even on its own — Cochrane found it lacked adequate trial data for conclusions, and the AAD guideline lists it under insufficient evidence. One controlled trial of a combined benzoyl peroxide and precipitated sulfur cream exists, from 1966, and its results are not retrievable from any index we can reach. There is no documented chemical incompatibility.
In practice
Both are drying and keratolytic, so additive irritation is the practical concern. There is no evidence promising you extra benefit from stacking them.
Sources
Nobody has tested this Benzoyl peroxide + Tea tree oil
In short
Untested, and there is a reason to be wary that has nothing to do with efficacy.
What the evidence says
No study has tested them together. The famous trial is a head-to-head comparison, not a combination study. The concern is that tea tree oil is a common contact allergen and its oxidised degradation products are substantially more sensitising than fresh oil — pairing a volatile, oxidisable essential oil with a strong oxidiser is chemically imprudent even though nobody has measured it.
Commonly repeated, and not quite right
Commonly repeated: that the tea tree oil trial shows it works alongside benzoyl peroxide. What it actually was: a head-to-head comparison of one against the other. It tells you nothing about combining them.
In practice
There is no evidence of added benefit and a plausible reason for added risk. If you want to try tea tree oil, try it on its own, from a fresh bottle.
Sources
Keep them apart Benzoyl peroxide + Physical scrub or cleansing brush
In short
Skip the scrub. This one is expert consensus rather than trial data, and we will say so.
What the evidence says
No controlled trial of scrubs with actives exists, and the 2024 guideline makes no recommendation on mechanical exfoliation at all. The AAD's patient guidance is the best available source and is specific: exfoliating while using retinoid creams, retinol or benzoyl peroxide may worsen dry skin or even cause breakouts, and those with acne-prone skin may find mechanical exfoliation too irritating.
In practice
The mechanism — barrier disruption compounding the disruption the active already causes — is sound, and the upside of a scrub over a chemical exfoliant is close to nil, so the risk-benefit is clear even without a trial.
Sources
Nobody has tested this Benzoyl peroxide + Clascoterone (Winlevi)
In short
No combination trial, but the guideline's whole logic points toward combining.
What the evidence says
Clascoterone blocks androgen receptors in the skin; benzoyl peroxide is an antimicrobial and comedolytic. Different mechanisms, no shared chemistry. The guideline's first good practice statement recommends multimodal topical therapy combining multiple mechanisms of action. What does not exist is a trial of this specific pair.
In practice
Reasonable to combine. Introduce one at a time so you know which is doing what — and which is causing any irritation.
Sources
Why most “don’t mix” rules are not what they look like
Three different things get flattened into the same warning, and separating them is most of the value here. A chemical interaction means one ingredient degrades or reacts with the other — rare, and when it is real there is usually a stability study or an FDA-approved product engineered around it. Additive irritation means nothing reacts; your skin barrier simply takes two hits instead of one. That is a dose-and-frequency problem, not an incompatibility, and it is fixed by spacing things out rather than by avoiding one of them forever. A prescribing interaction is a medical question about what happens inside your body, and it belongs with a prescriber or a pharmacist, not a web page.
The fourth category is the one nobody admits to: no evidence at all. Most ingredient pairs have never been studied together, because there is no commercial reason to run that trial. When you pick a combination we have not researched, this tool will say so plainly and then reason from what each ingredient is known to do — clearly labelled as reasoning rather than a finding, because those are not the same thing and it matters which one you are being handed.
What this cannot do
It does not know your skin, your medical history, what else you take, or whether you are pregnant. It cannot tell you whether two prescriptions are safe together — that is a question for the person who prescribed them, or for a pharmacist, who will check an interaction for you quickly and at no cost. And it is not a reason to start, stop or change anything. It is a way to find out what is actually known, so that the conversation you have with a clinician is a better one.
Related: the acne treatment evidence map — every treatment, with how strongly it is recommended and how certain the evidence is. Spot an error in either? Tell us; we log every correction on our corrections page.