Every acne treatment in one place, with the two things almost nobody publishes together: how strongly the guideline actually recommends it, and how certain the evidence behind it actually is. Those are different questions, and the gap between them is where most acne advice goes wrong.
This is built almost entirely from the American Academy of Dermatology’s 2024 acne guideline — 18 graded recommendations, 5 good practice statements, and a long list of things the panel looked at and could not make a call on. That last list is the most useful part and the part that never gets written up, so it is here in full. Where a row falls outside what that guideline covers — scars and pigmentation, mainly — it says so, and the rating comes from a named systematic review instead.
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.
| Treatment | Type | Guideline position | Certainty of evidence | How long it takes | Detail |
|---|---|---|---|---|---|
| Adapalene + benzoyl peroxide (fixed dose)OTC and prescription strengths | Over-the-counter topical | Strong recommendation | Moderate | 4–12 weeks | |
What the evidence actually saysA separate strong, moderate-certainty recommendation for fixed-dose topical retinoid with benzoyl peroxide. Adapalene is chemically stable next to benzoyl peroxide, which is why this combination exists as a single product. The catchMore irritating than either alone. Bleaches fabric. Pregnancy & breastfeedingAvoid in pregnancy because of the retinoid. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads, Back & chest |
|||||
| Adapalene 0.1% (OTC retinoid)Over the counter | Over-the-counter topical | Strong recommendation | Moderate | 8–12 weeks, often worse before better | |
What the evidence actually saysCovered by the guideline's strong, moderate-certainty recommendation for topical retinoids as a class. Adapalene is the only retinoid available without a prescription in the US and is the most photostable of the four. The catchPurging, dryness and peeling in the first 4–8 weeks are typical. The guideline gives no retinoid superiority over another. Pregnancy & breastfeedingAvoid in pregnancy — topical retinoids are not recommended. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Red inflamed spots, Keeping it clear |
|||||
| Benzoyl peroxideOver the counter | Over-the-counter topical | Strong recommendation | Moderate | 4–12 weeks | |
What the evidence actually saysOne of only two topical classes the AAD gives a strong recommendation with moderate-certainty evidence. It is also the only acne active with no documented Cutibacterium acnes resistance, which is why guidelines pair it with every antibiotic. Commonly repeated, and not quite rightCommonly repeated: that a higher percentage works better. What the evidence suggests: comparative work has generally found little efficacy difference between roughly 2.5% and 10%, while irritation rises with strength. The guideline recommends benzoyl peroxide without specifying a concentration. Starting low is a reasonable reading of that. The catchBleaches towels, pillowcases and hair. Drying at higher strengths — 2.5% works about as well as 10% for most people. Pregnancy & breastfeedingGenerally considered compatible with pregnancy; confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads, Back & chest, Keeping it clear |
|||||
| Clindamycin + benzoyl peroxide (fixed dose)Prescription | Prescription topical | Strong recommendation | Moderate | 6–12 weeks | |
What the evidence actually saysIts own strong, moderate-certainty recommendation, separate from the individual ingredients. The pairing is what prevents the antibiotic resistance that clindamycin alone causes. The catchBleaches fabric. Some formulations need refrigeration. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Back & chest Sources for this row |
|||||
| Clindamycin + tretinoin (fixed dose)Prescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks | |
What the evidence actually saysStrong, moderate-certainty recommendation for fixed-dose topical retinoid with topical antibiotic — with an explicit remark that concomitant benzoyl peroxide is recommended to prevent resistance. The catchContains no benzoyl peroxide, so the guideline says to add one separately for resistance control. Label warns against combining with erythromycin-containing products. Pregnancy & breastfeedingContraindicated in pregnancy because of the retinoid. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads |
|||||
| Oral doxycyclinePrescription | Oral medication | Strong recommendation | Moderate | 6–12 weeks; usually capped at 3–4 months | |
What the evidence actually saysThe only oral antibiotic with a strong, moderate-certainty recommendation, and conditionally preferred over azithromycin (low certainty). Two of the guideline's five good practice statements are about limiting systemic antibiotics and always pairing them with benzoyl peroxide and other topicals. The catchPhotosensitivity, oesophageal irritation, stomach upset. Must be paired with topicals and time-limited. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Deep / cystic, Back & chest |
|||||
| TazarotenePrescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks | |
What the evidence actually saysCovered by the strong, moderate-certainty class recommendation for topical retinoids. No drug-specific recommendation exists. The catchOften the most irritating of the four retinoids. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Red inflamed spots |
|||||
| Topical clindamycinPrescription | Prescription topical | Strong recommendation | Moderate | 6–12 weeks | |
What the evidence actually saysStrong, moderate-certainty recommendation for topical antibiotics as a class, with an explicit remark that monotherapy is not recommended. In one clinical comparison, clindamycin-resistant C. acnes counts rose to over 1600% of baseline on monotherapy but stayed at or below baseline when paired with benzoyl peroxide. Commonly repeated, and not quite rightCommonly repeated: that a topical antibiotic is a gentle first thing to try on its own. What the guideline actually says: the recommendation for topical antibiotics carries an explicit remark that 'topical antibiotic monotherapy is not recommended.' In one clinical comparison, clindamycin-resistant bacteria rose to over 1600% of baseline on clindamycin alone, but stayed at or below baseline when it was paired with benzoyl peroxide. The catchNever use it alone. Monotherapy breeds resistant bacteria — pair it with benzoyl peroxide every time. Pregnancy & breastfeedingOften used in pregnancy; confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots |
|||||
| Topical dapsonePrescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks | |
What the evidence actually saysFolded into the strong, moderate-certainty topical antibiotic class recommendation rather than graded separately. The guideline makes two practical points about it: no G6PD screening needed, and the benzoyl peroxide discoloration. The catchTurns skin and facial hair orange-brown if layered directly with benzoyl peroxide. G6PD screening is not required. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots |
|||||
| Topical erythromycinPrescription | Prescription topical | Strong recommendation | Moderate | 6–12 weeks | |
What the evidence actually saysWithin the strong, moderate-certainty topical antibiotic class recommendation. The guideline does not compare erythromycin against clindamycin directly. The catchHigher background resistance than clindamycin in most regions. Never use alone. Pregnancy & breastfeedingOften used in pregnancy; confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Topical minocycline foamPrescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks | |
What the evidence actually saysWithin the strong, moderate-certainty topical antibiotic class recommendation. FDA-approved from age 9. The catchExpensive. Same stewardship rule: pair with benzoyl peroxide. Pregnancy & breastfeedingAvoid — tetracycline class. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| TretinoinPrescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks, often worse before better | |
What the evidence actually saysCovered by the strong, moderate-certainty class recommendation for topical retinoids. The guideline states plainly that existing comparative data do not show one topical retinoid to be superior to another. Tretinoin also has separate randomised evidence for post-inflammatory hyperpigmentation in darker skin. The catchDegrades in light and next to benzoyl peroxide in older gel formulations. Purging is common. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Red inflamed spots, Keeping it clear, Dark marks |
|||||
| Tretinoin + benzoyl peroxide (encapsulated, Twyneo)Prescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks | |
What the evidence actually saysCovered by the strong, moderate-certainty recommendation for fixed-dose retinoid plus benzoyl peroxide. It is worth knowing why this product needed inventing: tretinoin and benzoyl peroxide genuinely do degrade each other in a shared film, so the two actives are separately micro-encapsulated in silica shells. The catchBleaches fabric. Retinoid irritation still applies. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads |
|||||
| TrifarotenePrescription | Prescription topical | Strong recommendation | Moderate | 8–12 weeks | |
What the evidence actually saysCovered by the strong, moderate-certainty class recommendation for topical retinoids. Its distinguishing feature is trial data on truncal acne, not superiority on the face. The catchNewest and usually the most expensive retinoid. Studied on the trunk as well as the face. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Red inflamed spots, Back & chest Sources for this row |
|||||
| Intralesional corticosteroid injectionIn clinic | In-clinic procedure | Good practice statement | — | 24–72 hours for one lesion | |
What the evidence actually saysA good practice statement rather than a graded recommendation: recommended as adjuvant therapy for larger papules or nodules, used judiciously, at lower concentration and volume to limit local steroid side effects. The catchCan leave a temporary dent or a pale patch. It treats one lesion, not the disease. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDeep / cystic Sources for this row |
|||||
| Oral isotretinoinPrescription, with a pregnancy prevention programme | Oral medication | Good practice statement | — | 4–6 months for a typical course; improvement often continues after stopping | |
What the evidence actually saysWorth reading carefully: isotretinoin's core recommendation is a good practice statement, not a graded strong recommendation — the panel considered the case so clear that formal grading was not applied. The same statement says patients with psychosocial burden or scarring should be treated as having severe acne, and that population-based studies have NOT identified an increased risk of neuropsychiatric conditions or inflammatory bowel disease. Commonly repeated, and not quite rightCommonly repeated: that isotretinoin has been shown to cause depression and inflammatory bowel disease. What the guideline actually says: 'Population-based studies have not identified increased risk of neuropsychiatric conditions or inflammatory bowel disease in acne patients undergoing treatment with isotretinoin.' That is a statement about population-level risk, and it is not the same as saying no individual is ever affected — mood changes are still worth raising with your prescriber promptly. It does mean the blanket claim is not what the current evidence shows. The catchAbsolutely contraindicated in pregnancy. Monitoring of liver function and lipids should be considered; routine blood-count monitoring is not needed in healthy patients. Avoid waxing and resurfacing during treatment and for six months after. Pregnancy & breastfeedingAbsolutely contraindicated. Pregnancy prevention is mandatory. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDeep / cystic, Red inflamed spots, Back & chest, Scars |
|||||
| Clascoterone 1% cream (Winlevi)Prescription | Prescription topical | Conditional recommendation | High | 12 weeks | |
What the evidence actually saysThis is the most misread row in the guideline. Clascoterone has HIGH certainty evidence — the highest of any acne topical — yet only a conditional recommendation. The guideline's own footnote says the downgrade was made because the cost 'may impact equitable acne treatment access'. It is the first topical androgen-receptor blocker, and the only one usable by men. Commonly repeated, and not quite rightCommonly repeated: that a 'conditional' recommendation means the evidence is shaky. What the guideline actually says: clascoterone has high certainty evidence — better than benzoyl peroxide's or tretinoin's, which are both moderate. The recommendation was softened to conditional for a reason the guideline states outright in a footnote: the cost 'may impact equitable acne treatment access.' That is a fairness judgement about access, not a doubt about whether the drug works. The catchThe conditional rating is about price, not performance. Listed at roughly $590–$671 per tube at the time of the guideline. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forHormonal pattern, Red inflamed spots |
|||||
| Oral sarecyclinePrescription | Oral medication | Conditional recommendation | High | 12 weeks | |
What the evidence actually saysHigh-certainty evidence, conditional recommendation. The guideline's footnote is explicit that the downgrade for sarecycline and clascoterone was made 'due to high current cost of treatment that may impact equitable acne treatment access'. Its narrower spectrum is the theoretical advantage. Commonly repeated, and not quite rightCommonly repeated: that sarecycline is conditionally recommended because it is less proven than doxycycline. What the guideline actually says: sarecycline's evidence certainty is high; doxycycline's is moderate. Doxycycline gets the stronger recommendation anyway, and the guideline's footnote attributes sarecycline's conditional rating to its cost. Both facts are worth holding at once. The catchLike clascoterone, downgraded to conditional on cost grounds rather than on evidence. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Deep / cystic Sources for this row |
|||||
| Azelaic acidOTC 10% and prescription 15–20% | Over-the-counter topical | Conditional recommendation | Moderate | 4–12 weeks; longer for marks | |
What the evidence actually saysConditional recommendation on moderate-certainty evidence. Its distinctive feature is that it also lightens post-inflammatory hyperpigmentation, which benzoyl peroxide and retinoids do not do directly. The catchCochrane found it probably works less well than benzoyl peroxide. Can sting for the first week or two. Pregnancy & breastfeedingOne of the few acne actives commonly used in pregnancy — still confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads, Dark marks |
|||||
| Combined oral contraceptive pillsPrescription | Oral medication | Conditional recommendation | Moderate | 3–6 months | |
What the evidence actually saysConditional recommendation on moderate-certainty evidence, with a footnote saying the downgrade is 'due to the variability in patient values and preferences related to contraception and hormonal medications' — a values judgement, not an evidence one. The catchThe conditional rating reflects variability in what people want from a contraceptive, not weak efficacy. Clotting and blood-pressure considerations apply. Pregnancy & breastfeedingNot applicable. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forHormonal pattern, Red inflamed spots Sources for this row |
|||||
| Oral minocyclinePrescription | Oral medication | Conditional recommendation | Moderate | 6–12 weeks | |
What the evidence actually saysConditional recommendation on moderate-certainty evidence — a step below doxycycline, which is strong. The side-effect profile is the reason most guidelines now lead with doxycycline. The catchRare but real risks doxycycline does not carry: drug-induced lupus, hyperpigmentation, vestibular side effects. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Deep / cystic Sources for this row |
|||||
| SpironolactonePrescription | Oral medication | Conditional recommendation | Moderate | 3–6 months | |
What the evidence actually saysConditional recommendation on moderate-certainty evidence, with a remark most people have not heard: 'Potassium monitoring is not needed in healthy patients.' That reflects direct research finding routine potassium checks of low usefulness in healthy young women. Testing is still appropriate with older age, kidney or heart disease, or other potassium-raising drugs. Commonly repeated, and not quite rightCommonly repeated: that you need regular potassium blood tests on spironolactone. What the guideline actually says: 'Potassium monitoring is not needed in healthy patients.' That language follows research that found routine potassium checks of low usefulness in healthy young women taking it for acne. Testing is still appropriate with older age, kidney or heart disease, or other medications that raise potassium — so this is a conversation with your prescriber, not a rule to apply yourself. The catchFetal risk means reliable contraception matters. Breast tenderness and irregular periods are common early on. Pregnancy & breastfeedingAvoid in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forHormonal pattern, Red inflamed spots |
|||||
| Salicylic acidOver the counter | Over-the-counter topical | Conditional recommendation | Low | 4–8 weeks | |
What the evidence actually saysConditionally recommended on low-certainty evidence. Cochrane found little or no difference between salicylic acid and tretinoin in treatment response, but rated that comparison low certainty too. The catchLow-certainty evidence. Better for blackheads and whiteheads than for deep inflamed lesions. Pregnancy & breastfeedingTopical use on limited areas is generally considered acceptable; confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Keeping it clear |
|||||
| Low-glycaemic-load dietSelf-directed | Diet & supplements | Evidence called conflicting | — | 10–12 weeks in the trials | |
What the evidence actually saysThe guideline reviewed four randomised trials: three favourable, one null. Rather than 'insufficient', it describes the evidence as conflicting and makes no recommendation. That is the most honest summary available of the diet question. Commonly repeated, and not quite rightCommonly repeated: that dermatology guidelines say diet has nothing to do with acne. What the guideline actually says: for low-glycaemic-load diets specifically, it describes the evidence as conflicting rather than insufficient — three of four randomised trials favourable, one null. 'Conflicting' is a different and more interesting word than 'insufficient', and the guideline uses it deliberately here and nowhere else. The catchThe guideline calls this evidence conflicting, not insufficient — a meaningful distinction. Pregnancy & breastfeedingNo concern with a balanced version. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads Sources for this row |
|||||
| Chemical peels for active acneIn clinic | In-clinic procedure | Evidence judged insufficient | — | A series of 3–6 sessions | |
What the evidence actually saysThe guideline names glycolic, trichloroacetic, salicylic, Jessner's and mandelic acid peels together in its insufficient-evidence statement for acne treatment. Systematic reviews of peels exist but did not meet the threshold the panel used. Commonly repeated, and not quite rightWorth separating: the guideline's insufficient-evidence finding is about peels for active acne. Peels for pigmentation and for scarring are a different question with a different (though still thin) literature. A source that cites the guideline against peels generally is over-reading it. The catchInsufficient evidence per the guideline — for active acne. Peels for pigmentation are a different question. Pregnancy & breastfeedingDiscuss with your clinician; some agents are avoided. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Red inflamed spots, Dark marks |
|||||
| Chocolate avoidanceSelf-directed | Diet & supplements | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement on diet. The most famous acne food rule has no guideline-grade evidence behind it. Commonly repeated, and not quite rightCommonly repeated: that chocolate causes breakouts, or alternatively that this has been definitively debunked. What the guideline actually says: the evidence is insufficient to make a recommendation either way. Neither confident claim is supported. The catchInsufficient evidence per the guideline, in either direction. Pregnancy & breastfeedingNo concern. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Comedo extractionIn clinic | In-clinic procedure | Evidence judged insufficient | — | Immediate for the lesion extracted | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement on procedural treatments. The catchInsufficient evidence per the guideline. Done badly it causes the scarring it is meant to avoid. Pregnancy & breastfeedingNo systemic concern. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads Sources for this row |
|||||
| FlutamidePrescription | Oral medication | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement. Rarely used for acne in current practice. The catchInsufficient evidence per the guideline, and it carries a risk of liver injury. Pregnancy & breastfeedingAvoid. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forHormonal pattern Sources for this row |
|||||
| Lasers and light devices for active acneIn clinic | In-clinic procedure | Evidence judged insufficient | — | A series of sessions | |
What the evidence actually saysThe guideline lists the devices it reviewed by wavelength — pulsed dye, Nd:YAG, 1450 diode, KTP, infrared LED, red light, blue-and-red light combinations, intense pulsed light — and concludes the evidence is insufficient for all of them for acne treatment. Note this is about active acne, not scars. Commonly repeated, and not quite rightWorth knowing: the 2024 guideline contains exactly one recommendation against anything, and it is a device — adding pneumatic broadband light to adapalene 0.3% gel (conditional, low certainty). For every other laser and light device it reviewed, by specific wavelength, it concluded the evidence was insufficient. Device marketing rarely mentions either finding. The catchInsufficient evidence per the guideline, across a long list of specific devices. Expensive, rarely covered by insurance. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Low-dairy or low-whey dietSelf-directed | Diet & supplements | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement on diet, alongside omega-3 fatty acids and chocolate. The catchInsufficient evidence per the guideline. Observational associations are not the same as a trial. Pregnancy & breastfeedingWatch calcium intake if you cut dairy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| MetforminPrescription | Oral medication | Evidence judged insufficient | — | Unclear for acne | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement, alongside oral corticosteroids and flutamide. Its role in PCOS is a separate question from its effect on acne lesion counts. The catchInsufficient evidence for acne specifically, even though it is widely used in PCOS. Pregnancy & breastfeedingA clinician decision. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forHormonal pattern Sources for this row |
|||||
| Omega-3 fatty acid supplementsOver the counter | Diet & supplements | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement on diet for acne treatment. The catchInsufficient evidence per the guideline. Pregnancy & breastfeedingGenerally considered safe; check with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Oral azithromycinPrescription | Oral medication | Evidence judged insufficient | — | Variable | |
What the evidence actually saysThe guideline found the evidence insufficient to recommend azithromycin on its own, and separately made a conditional, low-certainty recommendation for doxycycline over azithromycin. The catchDoxycycline is conditionally preferred over it, on low-certainty evidence. Pregnancy & breastfeedingSometimes used when a macrolide is needed in pregnancy; a clinician decision. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Oral corticosteroidsPrescription | Oral medication | Evidence judged insufficient | — | Days, for flares | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement for acne treatment. The catchInsufficient evidence per the guideline. Short-term rescue use only, and steroids can themselves cause an acneiform eruption. Pregnancy & breastfeedingA clinician decision. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDeep / cystic Sources for this row |
|||||
| Oral or topical zincOver the counter | Diet & supplements | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement. Cochrane reviewed zinc too and did not find a basis for confident conclusions. The catchInsufficient evidence per the guideline. High-dose oral zinc causes nausea and can affect copper absorption. Pregnancy & breastfeedingStay within recommended daily amounts. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots |
|||||
| Oral pantothenic acid (vitamin B5)Over the counter | Diet & supplements | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement for acne treatment. The catchInsufficient evidence per the guideline. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Photodynamic therapyIn clinic | In-clinic procedure | Evidence judged insufficient | — | A series of sessions | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement on procedural treatments, specifically photodynamic therapy with aminolevulinic acid. The catchInsufficient evidence per the guideline. Significant downtime and photosensitivity after treatment. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Deep / cystic Sources for this row |
|||||
| ResorcinolOver the counter | Over-the-counter topical | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement. It survives mainly as a legacy US OTC monograph ingredient, permitted only in combination with sulfur. The catchInsufficient evidence per the guideline. Largely superseded. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads |
|||||
| Sodium sulfacetamideOTC and prescription | Over-the-counter topical | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement. Long-standing clinical use, thin modern trial base. The catchInsufficient evidence per the guideline. Turns skin and facial hair orange when layered with benzoyl peroxide. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots |
|||||
| SulfurOver the counter | Over-the-counter topical | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement alongside glycolic acid, sodium sulfacetamide and resorcinol. Cochrane likewise found sulphur lacked adequate trial data for conclusions. The catchThe guideline found the evidence insufficient. Smells of sulfur, because it is sulfur. Pregnancy & breastfeedingGenerally considered low-risk topically; confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads |
|||||
| Tea tree oilOver the counter | Over-the-counter topical | Evidence judged insufficient | — | Slower than benzoyl peroxide in the one comparison trial | |
What the evidence actually saysNamed in the guideline's insufficient-evidence statement. The single most-cited study, Bassett 1990, compared 5% tea tree oil against 5% benzoyl peroxide: both improved acne, tea tree oil more slowly and with fewer side effects. One small single-blind trial is not a basis for a recommendation. Commonly repeated, and not quite rightCommonly repeated: that tea tree oil has been proven to work as well as benzoyl peroxide. What that study actually found: Bassett 1990 was a single-blind trial of 124 people comparing the two head to head. Both improved acne; tea tree oil worked more slowly and caused fewer side effects. One small single-blind trial is why the AAD guideline still lists tea tree oil under insufficient evidence — and that trial is also frequently mis-cited as evidence for using the two together, which it says nothing about. The catchInsufficient evidence per the guideline. A common contact allergen, and oxidised (old) tea tree oil is substantially more sensitising. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots |
|||||
| Topical glycolic acidOver the counter | Over-the-counter topical | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement for acne treatment. Most of its reputation comes from peel studies and from pigmentation rather than from acne lesion counts. The catchInsufficient evidence for acne per the guideline. Increases sun sensitivity. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forBlackheads & whiteheads, Dark marks |
|||||
| Topical green teaOver the counter | Over-the-counter topical | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement for acne. The catchInsufficient evidence per the guideline. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Topical niacinamideOver the counter | Over-the-counter topical | Evidence judged insufficient | — | 8–12 weeks if at all | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement. Smaller studies suggest benefit for barrier support and pigmentation, and a split-face trial found 5% niacinamide added to 2.5% benzoyl peroxide beat benzoyl peroxide alone — but this is not guideline-grade evidence. Commonly repeated, and not quite rightCommonly repeated: that niacinamide is an evidence-based acne active. What the guideline actually says: oral and topical niacinamide are both named in the insufficient-evidence statement. There is supportive smaller work — a split-face trial found 5% niacinamide added to benzoyl peroxide beat benzoyl peroxide alone — but that is a long way from guideline-grade evidence, and it is fair to say so. The catchThe guideline found the evidence insufficient to make any recommendation, for oral or topical. Pregnancy & breastfeedingNo known pregnancy concern. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Dark marks |
|||||
| Trimethoprim/sulfamethoxazolePrescription | Oral medication | Evidence judged insufficient | — | Variable | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement. It persists as a second- or third-line option in practice rather than on the strength of the acne evidence. The catchInsufficient evidence per the guideline. Carries a rare risk of severe cutaneous reactions. Pregnancy & breastfeedingAvoid — discuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Deep / cystic Sources for this row |
|||||
| Witch hazelOver the counter | Over-the-counter topical | Evidence judged insufficient | — | Unclear | |
What the evidence actually saysNamed explicitly in the guideline's insufficient-evidence statement for acne. The catchInsufficient evidence per the guideline. Alcohol-containing formulations can be drying. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Azelaic acid for post-inflammatory hyperpigmentationOTC 10% and prescription 15–20% | Scars & marks | Not assessed in the guideline | Moderate | 8–16 weeks | |
What the evidence actually saysHas direct trial evidence for post-inflammatory hyperpigmentation as well as for acne itself, which is a rare combination among acne actives. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Pregnancy & breastfeedingCommonly used in pregnancy; confirm with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDark marks |
|||||
| Fractional laser resurfacing for acne scarsIn clinic | Scars & marks | Not assessed in the guideline | Moderate | 3–5 sessions; months of remodelling after | |
What the evidence actually saysA network meta-analysis compared six photoelectric therapies for atrophic acne scars head to head — one of the better evidence bases in scar treatment, though still built on small trials. Non-ablative fractional resurfacing in skin of colour has its own evidence-based review. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Downtime, cost, and a real risk of post-inflammatory hyperpigmentation in skin of colour — non-ablative settings and an experienced operator matter. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. |
|||||
| Sunscreen for post-inflammatory hyperpigmentationOver the counter | Scars & marks | Not assessed in the guideline | Moderate | Prevention; months to see the difference | |
What the evidence actually saysThe least glamorous and best-supported thing you can do for dark marks: UV and visible light both drive the pigment that makes them last. Iron-oxide-containing formulations have direct evidence against visible-light-induced pigmentation in skin of colour. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Tinted sunscreens containing iron oxides also block visible light, which matters in darker skin tones. Pregnancy & breastfeedingNo concern with mineral filters. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDark marks |
|||||
| Topical tretinoin for post-inflammatory hyperpigmentationPrescription | Scars & marks | Not assessed in the guideline | Moderate | 10–40 weeks | |
What the evidence actually saysHas a randomised controlled trial specifically in Black patients with post-inflammatory hyperpigmentation — unusually good evidence for a pigmentation treatment, and in the population where it matters most. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Retinoid irritation can itself cause more pigmentation in darker skin, so start slow. Pregnancy & breastfeedingContraindicated in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDark marks Sources for this row |
|||||
| Hydroquinone (with tretinoin and a steroid, Kligman-type)Prescription | Scars & marks | Not assessed in the guideline | Low | 8–12 weeks, short-term use | |
What the evidence actually saysThe classic triple formula dates to Kligman and Willis in 1975 and there is an FDA-approved product for melasma. But a 2025 meta-analysis found no statistically significant advantage over alternative treatments on standard pigmentation scores, with high heterogeneity — and the melasma evidence does not transfer automatically to acne marks. Commonly repeated, and not quite rightCommonly repeated: that the Kligman triple formula is the gold standard for pigmentation. What a 2025 meta-analysis found: no statistically significant advantage over alternative treatments on standard pigmentation scores, with high heterogeneity between studies. It remains widely used and FDA-approved for melasma; 'best available and well established' is a fairer description than 'gold standard', and the melasma evidence does not transfer automatically to acne marks. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Short-term, clinician-supervised only. Prolonged use risks exogenous ochronosis — a blue-black darkening that should prompt stopping. Do not layer with benzoyl peroxide. Pregnancy & breastfeedingAvoid in pregnancy. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forDark marks |
|||||
| Microneedling for atrophic acne scarsIn clinic | Scars & marks | Not assessed in the guideline | Low | 3–6 sessions, months apart | |
What the evidence actually saysA systematic review and meta-analysis of microneedling monotherapy found benefit for atrophic scars, with the usual caveats of small trials and subjective scar scales. Adding platelet-rich plasma has been studied and reviewed separately. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Expect improvement, not erasure. Higher risk of post-inflammatory pigmentation in darker skin. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forScars |
|||||
| Subcision for tethered (rolling) scarsIn clinic | Scars & marks | Not assessed in the guideline | Low | 1–3 sessions | |
What the evidence actually saysEvidence is mostly small comparative studies rather than large randomised trials, often combined with microneedling. It is the one technique that targets the fibrous tether pulling the scar down, which no topical can reach. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Bruising is expected. Works on tethered rolling scars, not on ice-pick scars. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forScars |
|||||
| TCA CROSS for ice-pick scarsIn clinic | Scars & marks | Not assessed in the guideline | Very low | 3–6 sessions, weeks apart | |
What the evidence actually saysWidely used for narrow deep scars that lasers cannot reach, but the published evidence is largely case series rather than controlled trials. The catchThe 2024 AAD acne guideline covers acne treatment, not scar or pigmentation treatment. There is no equivalent graded guideline for these, so the evidence rating here is our reading of the named systematic review, not a guideline position. Deliberately injures the scar base. Operator skill matters enormously; pigmentation risk in darker skin. Pregnancy & breastfeedingDiscuss with your clinician. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forScars Sources for this row |
|||||
| Clindamycin + adapalene + benzoyl peroxide (triple, Cabtreo)Prescription | Prescription topical | Not assessed in the guideline | — | 8–12 weeks | |
What the evidence actually saysNot addressed by the 2024 guideline — recommendations 1.8 to 1.10 cover two-drug fixed-dose combinations only. Each of its three components is individually recommended, but the specific triple product was not reviewed. That is an absence of assessment, not a negative finding. The catchPostdates the guideline's evidence review, so it has no graded recommendation at all. Pregnancy & breastfeedingContraindicated in pregnancy because of the retinoid. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots, Blackheads & whiteheads, Back & chest Sources for this row |
|||||
| Hydrocolloid pimple patchesOver the counter | Over-the-counter topical | Not assessed in the guideline | — | 6–24 hours per lesion | |
What the evidence actually saysThe AAD guideline does not address them. The plausible mechanism — occlusive moist-wound dressing plus a physical barrier against picking — is well established in wound care, but there is no acne trial evidence of the quality the guideline requires. The catchNot assessed by the guideline. They manage one existing spot; they do not treat acne. Pregnancy & breastfeedingNo systemic exposure. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
| Moisturiser and broad-spectrum sunscreenOver the counter | Over-the-counter topical | Not assessed in the guideline | — | Supportive, not curative | |
What the evidence actually saysNot a graded recommendation, because it is adjunctive. There is direct trial evidence that adding a cleanser and an SPF 30 moisturiser to benzoyl peroxide 5% did not blunt its antibacterial effect, and reduced irritation. Sunscreen is also the single best-evidenced way to stop post-inflammatory marks from darkening. The catchNot a treatment. It is what lets you keep using the treatments that work. Pregnancy & breastfeedingNo concern with mineral filters; discuss chemical filters if you prefer. Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forKeeping it clear, Dark marks |
|||||
| Pneumatic broadband light added to adapalene 0.3%In clinic | In-clinic procedure | Conditional recommendation AGAINST | Low | n/a | |
What the evidence actually saysA conditional recommendation AGAINST, on low-certainty evidence. It is the single negative recommendation in the 2024 guideline, and worth knowing precisely because device marketing rarely mentions it. The catchThe only thing in the entire guideline recommended against. Pregnancy & breastfeedingn/a Always confirm with your own clinician — this is a summary of labelling, not advice for you. Used forRed inflamed spots Sources for this row |
|||||
No treatment matches those filters.
How to read the two ratings
These come from the American Academy of Dermatology’s 2024 acne guideline, which uses the GRADE system. The two columns answer different questions, and mixing them up is the single most common mistake made about this guideline.
- Strong recommendation
- The panel judged that almost everyone should have this option. Fourteen entries here.
- Good practice statement
- So clearly sensible the panel did not put it through formal grading. Isotretinoin’s core recommendation is one of these.
- Conditional recommendation
- Reasonable for many people, but the right choice depends on circumstances — including cost. Two entries here are conditional purely because of price, despite having the best evidence on the page.
- Recommendation against
- There is exactly one of these in the whole guideline.
- Insufficient / conflicting
- The panel looked and could not make a call. That is not the same as “it does not work” — it means the studies needed to know either way have not been done well enough.
- Not assessed
- Outside what the guideline covered, such as scar and pigmentation treatment, or newer than its evidence review. Ratings on those rows are our reading of the named systematic review, not a guideline position.
- Certainty: High
- Further research is very unlikely to change the estimate of the effect.
- Moderate
- Further research may well change it.
- Low / Very low
- Any estimate of the effect is uncertain. Most of what the internet is confident about lives here.
Where this comes from, and where it stops
The 2024 AAD guideline is the single best evidence synthesis available for acne treatment in the United States, and it is behind a paywall, written for clinicians, and thirty pages long. This page is an attempt to make its actual findings readable without softening them. Every strength and certainty rating in the table above is the guideline’s own, not our interpretation of it. Where we have added something — a timescale, a practical caveat, a note about cost — it sits in the detail panel, not in the rating.
Three things worth knowing before you use it. First, “conditional” does not mean “weak”: two of the treatments with the best evidence on this page are only conditionally recommended, and the guideline says plainly that the downgrade was about price, not performance. Second, “insufficient evidence” does not mean “doesn’t work” — it means the studies that would settle it have not been done well enough, which is a statement about the research, not about the treatment. Third, the guideline covers acne, not acne scarring; there is no equivalent graded guideline for scars or dark marks, and the rows covering those are flagged accordingly.
Found a mistake?
Several rows carry a note flagging a claim that gets repeated widely and does not survive contact with the source. We are not immune to that, and if you spot something here that is wrong, please tell us — we log every correction publicly on our corrections page with the date and what changed, and we write to the original source when the error is theirs rather than ours.
Next: can I use these two ingredients together? — the same treatment, applied to the combinations people actually ask about.