The American Academy of Dermatology’s 2024 acne guideline contains 23 numbered recommendations. Secondary summaries routinely describe them incorrectly, most often by reporting a recommendation as strong when the guideline never graded it at all, or by treating “strong recommendation” as a claim about evidence quality. This page lists all 23 as the guideline itself grades them, quotes what the guideline says it cannot recommend on, and links to the source so you can check every line.
Strength and certainty are two different things
This is the single most common source of error in acne content, and it is worth being precise about. Under the GRADE system the guideline uses, every graded recommendation carries two independent ratings:
- Strength: strong or conditional. This is about how confident the panel is that the desirable effects outweigh the undesirable ones for most patients. A strong recommendation means “most people should get this.”
- Certainty of evidence: high, moderate, low, or very low. This is about how much the underlying research can be trusted.
A strong recommendation can rest on moderate or even low-certainty evidence, and a merely conditional recommendation can rest on high-certainty evidence. In this guideline, both of those things happen, and the pattern runs in the direction most people would not guess.
The inversion. All three of the guideline’s high-certainty recommendations are only conditional. Every one of its seven strong recommendations rests on moderate certainty. Not a single strong recommendation in the 2024 acne guideline is backed by high-certainty evidence. When a page tells you a treatment is “strongly recommended by the AAD with high-quality evidence,” that combination does not exist in this document.
The strength × certainty grid
All 18 graded recommendations, placed on both axes at once. Numbers are the guideline’s own recommendation numbers.
| High certainty | Moderate certainty | Low certainty | |
|---|---|---|---|
| Strong 7 | none | 1.2 Benzoyl peroxide 1.3 Topical retinoids 1.4 Topical antibiotics 1.8 Topical antibiotic + BP 1.9 Topical retinoid + antibiotic 1.10 Topical retinoid + BP 2.1 Doxycycline |
none |
| Conditional, for 10 | 1.5 Clascoterone 2.3 Sarecycline 4.3 Standard or lidose isotretinoin |
1.7 Azelaic acid 2.2 Minocycline 3.1 Combined oral contraceptives 3.2 Spironolactone |
1.6 Salicylic acid 2.4 Doxycycline over azithromycin 4.2 Daily over intermittent isotretinoin |
| Conditional, against 1 | none | none | 5.1 Pneumatic broadband light added to adapalene 0.3% |
Certainty across all 18 graded recommendations: 3 high, 11 moderate, 4 low, 0 very low.
All 18 graded recommendations
| No. | Recommendation | Strength | Certainty |
|---|---|---|---|
| 1.2 | Benzoyl peroxide | Strong | Moderate |
| 1.3 | Topical retinoids | Strong | Moderate |
| 1.4 | Topical antibiotics | Strong | Moderate |
| 1.5 | Clascoterone | Conditional | High |
| 1.6 | Salicylic acid | Conditional | Low |
| 1.7 | Azelaic acid | Conditional | Moderate |
| 1.8 | Fixed-dose topical antibiotic + benzoyl peroxide | Strong | Moderate |
| 1.9 | Fixed-dose topical retinoid + antibiotic | Strong | Moderate |
| 1.10 | Fixed-dose topical retinoid + benzoyl peroxide | Strong | Moderate |
| 2.1 | Doxycycline (systemic) | Strong | Moderate |
| 2.2 | Minocycline (systemic) | Conditional | Moderate |
| 2.3 | Sarecycline (systemic) | Conditional | High |
| 2.4 | Doxycycline in preference to azithromycin | Conditional | Low |
| 3.1 | Combined oral contraceptive pills | Conditional | Moderate |
| 3.2 | Spironolactone | Conditional | Moderate |
| 4.2 | Traditional daily isotretinoin over intermittent dosing | Conditional | Low |
| 4.3 | Standard or lidose isotretinoin (equivalent options) | Conditional | High |
| 5.1 | Pneumatic broadband light added to adapalene 0.3% gel | Conditional against | Low |
The five good practice statements
These five are not graded. Under GRADE, a good practice statement is issued by expert consensus where the panel judged the benefit self-evident and a formal evidence review impractical or unnecessary. They carry no strength rating and no certainty rating.
| No. | Good practice statement | Grading |
|---|---|---|
| 1.1 | Multimodal topical therapy, combining mechanisms of action | Ungraded |
| 2.5 | Limiting the duration of systemic antibiotic use | Ungraded |
| 2.6 | Concomitant benzoyl peroxide with systemic antibiotics | Ungraded |
| 3.3 | Intralesional corticosteroid injection for individual lesions | Ungraded |
| 4.1 | Isotretinoin for severe acne | Ungraded |
Isotretinoin is the one to watch. Recommendation 4.1 is the statement most often misreported. It is repeatedly described across the web as a strong recommendation for severe acne. In the body of the guideline it is a good practice statement: ungraded, consensus-based, with no strength or certainty rating attached. This is not a criticism of isotretinoin, which is the most effective acne drug there is. It is a point about what the document actually says, and about how an error propagates when everyone summarises the summary instead of reading the source.
What the guideline says it cannot recommend on
The guideline contains six separate statements, one per section, declaring the evidence insufficient to make any recommendation. Each opens with the same formula: “Available evidence is insufficient to develop a recommendation on the use of…” These are the most misquoted lines in the document, usually by being converted into a recommendation against, which they are not. Insufficient evidence means the question was asked and could not be answered. It is not a verdict.
1: Topical therapies
Insufficient evidence to recommend on:
- Topical glycolic acid
- Sulfur
- Sodium sulfacetamide
- Resorcinol
The same statement also declares the evidence insufficient to compare topical benzoyl peroxide, retinoids, antibiotics and their combinations directly against each other. In other words: the guideline does not rank the topicals it strongly recommends.
2: Systemic antibiotics
Insufficient evidence to recommend on:
- Oral azithromycin
- Oral trimethoprim/sulfamethoxazole
Note that azithromycin still appears in graded recommendation 2.4, where doxycycline is conditionally preferred over it. Both statements are in the guideline and they are not in conflict.
3: Hormonal agents
Insufficient evidence to recommend on:
- Oral corticosteroids
- Flutamide
- Metformin
4: Physical modalities
Insufficient evidence to recommend on:
- Acne lesion and comedo extraction
- Chemical peels, including glycolic acid, trichloroacetic acid, salicylic acid, Jessner’s solution and mandelic acid
- Laser and light-based devices, including 585–595 nm pulsed dye laser, Nd:YAG laser, 1450 nm diode laser, potassium titanyl phosphate laser, infrared light-emitting diode, 635–670 nm red light, combined 420 nm blue and 660 nm red light, 589/1319 nm laser, and intense pulsed light
- Microneedle radiofrequency devices
- Photodynamic therapy with aminolevulinic acid
This is the longest of the six and the one most often misrepresented. Every laser, light device and peel on that list is in the “we could not answer this” category, not the “does not work” category.
5: Complementary and alternative therapies
Insufficient evidence to recommend on:
- Topical tea tree oil
- Topical green tea
- Topical witch hazel
- Oral pantothenic acid
- Oral or topical zinc
- Oral or topical niacinamide
6: Diet
Insufficient evidence to recommend on:
- Low dairy diet
- Low whey diet
- Omega-3 fatty acids
- Chocolate
Glycolic acid appears twice. Once in statement 1, as a topical agent, and again in statement 4, as a chemical peel. If you are citing the guideline on glycolic acid, say which one you mean.
What the guideline does not address at all
Some things appear in neither the graded recommendations nor any of the six insufficient-evidence statements. They may be discussed in the guideline’s narrative text, but no recommendation attaches to them, so there is no AAD position on them to cite. Any page attributing an AAD recommendation to the following is inventing it:
- Low-glycaemic-load diet as a treatment recommendation
- Probiotics, oral or topical
- Hydroquinone
- Over-the-counter “acne supplements” and detox products
- Blue and red light devices generally, other than the specific combinations named in statement 4 and in recommendation 5.1
One treatment, and only one, is recommended against. Recommendation 5.1: adding pneumatic broadband light to adapalene 0.3% gel: conditional against, low certainty. Nothing else in the 2024 guideline carries a recommendation against it. If you read that the AAD “advises against” dairy, tea tree oil, zinc, peels or lasers, that is a misreading of an insufficient-evidence statement.
Common questions
Does the AAD recommend against tea tree oil for acne?
No. Tea tree oil appears in the guideline’s complementary and alternative therapies section, in a statement that available evidence is insufficient to develop a recommendation on its use. That is a statement about the state of the research, not a judgement that it does not work. The guideline makes no recommendation either way.
Is isotretinoin strongly recommended by the AAD?
Not in the technical sense. Isotretinoin for severe acne is recommendation 4.1, a good practice statement: ungraded, based on panel consensus, with no strength or certainty rating. Two graded recommendations do concern isotretinoin: 4.2, conditionally favouring daily over intermittent dosing on low-certainty evidence, and 4.3, conditionally treating standard and lidose formulations as equivalent on high-certainty evidence.
Which acne treatments have the strongest evidence behind them?
Measured by certainty of evidence rather than strength of recommendation, three: clascoterone (1.5), sarecycline (2.3), and the equivalence of standard and lidose isotretinoin (4.3). All three are graded high certainty, and all three are only conditional recommendations. No recommendation in the guideline is both strong and high certainty.
Does the AAD say dairy causes acne?
No. The diet section states that available evidence is insufficient to develop a recommendation on low dairy diets, low whey diets, omega-3 fatty acids or chocolate. The guideline neither endorses nor rejects a dairy connection; it declines to rule.
How many recommendations are in the 2024 AAD acne guideline?
Twenty-three numbered recommendations: 18 formally graded under GRADE and 5 ungraded good practice statements. Of the 18 graded, 7 are strong, 10 are conditional in favour, and 1 is conditional against.
Source. Reynolds RV, Yeung H, Cheng CE, et al. Guidelines of care for the management of acne vulgaris. Journal of the American Academy of Dermatology. 2024;90(5):1006.e1–1006.e30. PMID 38300170.
Every figure on this page was coded directly from the guideline’s own recommendation tables rather than from any secondary summary, and re-checked against the full text. If you find an error, tell me and I will correct it and log the correction on the corrections page.
Keep reading
- The evidence map: 58 acne treatments, what the evidence says about each, and how many have none worth the name
- The ingredient combination checker: what can be layered with what
Not medical advice
This page describes what a clinical guideline says. It is not a treatment recommendation and it is not a substitute for a clinician who can examine your skin, know your history, and take responsibility for the decision. Guidelines describe what is reasonable for populations; your dermatologist decides what is reasonable for you.