Retinol for Asian Skin: Start Without Irritation

The safe way to start retinol as a beginner with Asian skin is to go low and slow. Pick a low strength, use it just once or twice a week at night to begin with, layer it with a good moisturiser, and wear sunscreen every morning. That gentle start is the whole game, because the biggest risk with retinol on Asian skin is not the ingredient itself but the irritation that comes from rushing it. And on skin that pigments easily, irritation can leave dark marks. Do it right, though, and retinol is one of the best things you can do for smoother, clearer, more even skin.

What is retinol and what does it do?

Retinol is a form of vitamin A, and it belongs to a family of ingredients called retinoids. Its main job is to speed up how quickly your skin cells renew themselves. That faster turnover is what smooths rough texture, unclogs pores, fades dark marks, and softens early fine lines over time. It is one of the most heavily researched skincare ingredients out there, with decades of evidence behind it.

There is a whole ladder of retinoids, from gentle over-the-counter retinol up to prescription tretinoin, which is more potent because it does not need to be converted by your skin first. For a beginner, plain retinol is the sensible starting rung. If retinol still feels like too much, a gentler plant-based alternative called bakuchiol exists. Bluemke and colleagues, publishing in the International Journal of Cosmetic Science in 2022, found that 23% of their 52-person panel reported reactions to a retinol formulation, things like redness, flaking, dryness and itching, while the bakuchiol formulation produced just one adverse reaction across the entire study. Worth knowing that research came from a skincare manufacturer, so hold it loosely, but the tolerability gap is hard to ignore. Our guide on bakuchiol versus retinol compares the two.

Why does Asian skin need extra care with retinol?

Asian skin is not more fragile, but it does react in a way worth respecting. The main issue is that retinoids can cause what dermatologists call retinoid dermatitis, meaning dryness, flaking, and redness while your skin adjusts. On skin that pigments easily, that irritation can then trigger post-inflammatory hyperpigmentation, the dark marks we cover in our hyperpigmentation guide.

Until recently that was mostly clinical impression rather than measured fact. Then Chang and colleagues actually counted. Working from 15 years of electronic medical records at Beth Israel Deaconess Medical Center, covering 2006 to 2020, they identified 1,456 adults prescribed topical retinoids for acne who self-identified as White, Black, or Asian. From those they built three matched cohorts of 251 people each, 753 in total, matched on age, sex, and the specific retinoid prescribed. Then they went through the clinical notes for documented intolerability, meaning burning, itching, redness, dryness, or peeling. The results were published in JAAD International in 2024.

Here are the raw numbers. Intolerability turned up in 31 of the 251 Asian patients, or 12.4 percent. In the White cohort it was 22 of 251, or 8.8 percent. In the Black cohort it was 8 of 251, or 3.2 percent. So the Asian group was the most affected of the three. In the multivariate analysis, using the Asian cohort as the reference point, the Black cohort had an odds ratio of 0.25 (95% CI 0.11 to 0.57, P = .001), which is statistically significant. The White cohort came in at 0.62 (95% CI 0.34 to 1.11, P = .109), which is a trend in the same direction but not significant. They also found women documented intolerability more often than men (P = .043).

Worth keeping this in proportion. Roughly one in eight Asian patients had irritation serious enough for a doctor to write it down, which means seven in eight did not. This is not a warning to stay away from retinoids. It's a reason to expect a bumpier first month than the average product review suggests, and to build your routine accordingly.

So if retinol has felt harsher on your skin than the reviews online suggested, that isn't you being dramatic or doing it wrong. It shows up in the medical record across hundreds of matched patients. And it's a reason to plan for a gentler start, not a reason to skip the ingredient.

Callender and colleagues flagged the mechanism in a clinical review of topical retinoids in skin of colour, published in the American Journal of Clinical Dermatology in 2022. It notes the risk of retinoid dermatitis in deeper skin tones and recommends starting at lower concentrations and choosing a cream or lotion over a gel to reduce irritation. So the extra care is not about avoiding retinol. It is about starting it in a way that keeps your skin calm, which protects you from the very pigmentation you are probably trying to fix.

How do you start retinol as a beginner?

Ease in over a couple of months rather than diving straight to nightly use. A gentle, well-tested schedule looks like this.

• Weeks 1 to 2: apply a pea-sized amount once a week, at night, to clean dry skin.
• Weeks 3 to 4: move to twice a week if your skin feels fine.
• Weeks 5 to 6: try every other night.
• Week 8 and beyond: settle at whatever frequency your skin tolerates comfortably, which for many people is two to four nights a week.

That schedule isn't something we made up. The American Academy of Dermatology tells beginners to use the least intense retinoid formula they can find, apply it every other night to start, and build up slowly, adding that people who skip that step "end up irritating their skin." The AAD also flags specifically that in deeper skin tones, irritation can trigger dark marks. Same two ideas you'll find on this page, from the largest dermatology body in the world.

Two practical details the AAD is firm about. Retinol goes on at night only, and you wear sun protection during the day, which means shade and a hat as well as sunscreen, not sunscreen alone. Neither is optional once retinol is in your routine.

This slow build is not just caution for its own sake. Remember that Bluemke panel: nearly a quarter of people using a retinol formulation reported redness, flaking or itching. Those reactions are dose and frequency dependent, so the same strength causes far less trouble when you introduce it gradually than when you apply it every night from day one. Patience genuinely changes how your skin responds.

What strength should beginners use?

Lower than you think. For a first retinol, something around 0.025% to 0.05%, or a gentle over-the-counter retinol serum, gives you real results while letting your skin adapt. The most common beginner mistake is reaching for the highest percentage on the shelf because more sounds better. It is not.

And this is not just caution. There is a trial that settles it. Griffiths and colleagues at the University of Michigan ran a 48-week double-blind, vehicle-controlled comparison of 0.1% and 0.025% tretinoin creams in 99 photoaged patients, published in Archives of Dermatology in 1995. Both strengths produced statistically significant improvement in photoaging versus vehicle, and there were no clinically or statistically significant differences in efficacy between them. Under the microscope the two looked nearly identical too, with epidermal thickening of 30% on 0.1% against 28% on 0.025%, and increased vascularity of 100% against 89%.

The one place they clearly differed was irritation. Erythema and scaling were significantly greater on 0.1%. So the higher strength bought no extra benefit over a year and cost noticeably more redness and flaking. That is a strong argument for starting low and staying there if it is working, and it matters double on skin that pigments in response to inflammation.

One fair objection to that trial: it tested tretinoin, which is prescription strength, not the retinol you buy over the counter. So does the same pattern hold for retinol itself? It does. Zasada and colleagues ran a split-face trial on 37 healthy women with a mean age of 46, applying 0.3% retinol serum to one cheek and 0.5% to the other for 12 weeks, published in Skin Pharmacology and Physiology in 2020. Same face, same routine, same 12 weeks, only the strength differed.

The result mirrors the tretinoin trial. Hyperpigmentation improved about 23% on the 0.5% side and 19% on the 0.3% side, and hydration actually improved more on the weaker side, 26% against 18%. The authors concluded there were no clinically or statistically significant efficacy differences between the two. But peeling, irritation, and itching were more frequent and more intense on the 0.5% cheek, and the volunteers themselves reported being happier with the lower strength.

Two trials, two different molecules, thirty years apart, same answer. Doubling or quadrupling the percentage buys irritation, not results. If you take one thing from this page, take that.

How do retinol percentages compare to prescription tretinoin?

This trips people up constantly, because 0.5% retinol sounds ten times stronger than 0.05% tretinoin when it's actually the opposite. Retinol has to be converted by your skin into retinoic acid before it does anything, and that conversion is inefficient. The working figure used in the research is that retinol is roughly ten times less potent than tretinoin. When researchers built retinol products to match prescription strengths for a split-face trial published in the Journal of Drugs in Dermatology, they formulated 0.25%, 0.5% and 1.0% retinol to correspond to 0.025%, 0.05% and 0.1% tretinoin. That ten-to-one ratio is the number to keep in your head when you're comparing labels.

Read that table alongside the Griffiths trial and the point lands hard. The top prescription strength beat the bottom one on irritation and matched it on results. So the bottom rows of this table aren't a consolation prize you graduate out of. For a lot of people they're the destination.

Texture matters too. As that skin of colour review notes, a cream or lotion formula tends to be gentler than a gel, which is worth knowing if your skin runs sensitive. There is also good evidence that retinol plays well with soothing partners. A 2025 study in the journal Cosmetics followed 43 Chinese women with acne-prone skin, including sensitive skin, for four weeks. Daily use of a product combining retinol, niacinamide and ceramides significantly cut both inflammatory and non-inflammatory lesion counts, reduced redness, and raised skin brightness. No adverse events came up across the whole study. The lesson is that formulation and support ingredients matter as much as the retinol percentage.

Which type of retinoid should a beginner pick?

Plain retinol or retinaldehyde, in a cream or lotion base. You can safely ignore most of the rest of the shelf. But since that shelf has got a lot more confusing lately, here's what the labels actually mean.

Walk into a Watsons, Guardian or Olive Young now and it isn't just percentages. You'll see retinal, granactive retinoid, HPR, encapsulated retinol, supramolecular retinol. Most of that is delivery-system marketing wrapped around a handful of real molecules.

How does the retinoid ladder actually work?

Your skin can only use one form, retinoic acid. Everything else has to be converted into it first, and the chain runs like this: retinyl esters become retinol, retinol becomes retinaldehyde, retinaldehyde becomes retinoic acid.

Each conversion step loses potency. So the further down the chain you buy, the gentler and slower the product. Prescription tretinoin is retinoic acid itself, which needs no conversion at all, which is exactly why it works fast and why it stings.

The head-to-head that settles the tolerability question is old but still the cleanest one there is. Fluhr and colleagues compared retinol, retinaldehyde and retinoic acid directly in Dermatology in 1999, running a 14-day patch test alongside a 44-week clinical study on 355 people.

Retinol and retinaldehyde came out with equally low irritation potential. Retinoic acid was significantly more irritating (P less than 0.05). In the first four weeks specifically, retinoic acid produced redness in 44 percent of the group, scaling in 35 percent, and burning or itching in 29 percent, while retinaldehyde produced all three significantly less often (P less than 0.0001). Laser Doppler readings backed up the same split.

One detail cuts the other way, and it's worth knowing. Both retinaldehyde and retinoic acid caused more scaling than plain retinol did (P less than 0.05). So retinal isn't a free lunch. It sits one conversion step from active, which means it works faster than retinol at a comparable strength while tolerating roughly as well, but you may flake a bit more. That makes it a perfectly reasonable first retinoid rather than something you have to graduate into.

If you do go that route, one recent trial gives you a strength to aim at, and it complicates the start-low rule in an honest way. Deda and colleagues ran a 24-week split-face, participant-blinded study on 56 women with a mean age of 44, comparing stabilised retinaldehyde at 0.1% against 0.05%, published in Skin Research and Technology in 2026. Both strengths raised dermal density significantly, 45.4 percent on the 0.1% side against 35.5 percent on the 0.05% side at the forehead, and both improved elasticity across every area measured. But only the 0.1% side produced statistically significant wrinkle reduction. The 0.05% side didn't move that measure at all.

So retinaldehyde doesn't behave quite like retinol here, and it's worth saying that plainly rather than flattening every finding into one rule. The useful detail is that the higher strength didn't cost much comfort. Just 1 of the 56 women reported significant irritation, and 85 percent said they could see an improvement. Read alongside the Fluhr data, that suggests 0.1% retinal is a sensible target to work up to rather than a strength to fear. Work up to it, though. Starting there in week one on skin that has never seen a retinoid is still how people end up with a peeling face and a fresh set of dark marks.

Are encapsulated and granactive retinoids worth paying more for?

Maybe, and there's regional data on both, but it doesn't say what the packaging implies.

Ye and colleagues tested a 0.1 percent cyclodextrin-encapsulated retinol essence, the thing usually sold as supramolecular retinol, on 32 middle-aged Chinese women over eight weeks, publishing in Skin Health and Disease in 2023. Stratum corneum hydration rose significantly by week four and kept climbing to week eight, elasticity measures improved by around 11 percent, pore grading improved 5.7 percent, and no adverse reactions were recorded across the entire study.

Shen and colleagues tested a serum pairing retinol with hydroxypinacolone retinoate, the HPR you'll see on labels as granactive retinoid, on 31 middle-aged Chinese women over eight weeks in the Journal of Cosmetic Dermatology in 2026. Elasticity rose 23.5 percent, hydration 56 percent, and pore volume dropped 57.2 percent. No participant had more than mild, transient dryness or redness.

Now the caveat, because it matters more than either set of numbers. Both studies are small, both ran only eight weeks, and crucially both were single-arm with no comparison group. Nobody put these formulations up against ordinary retinol. So what they show is that these products were well tolerated on Chinese skin, which is genuinely useful and not nothing. What they do not show is that encapsulated retinol or HPR beats plain retinol, because that test wasn't run. The HPR product's concentrations are proprietary too, so you can't even work out what dose produced the result.

The short version: the two things you can buy over the counter, retinol and retinal, are also the two gentlest options in the only proper head-to-head anyone has run. The encapsulation and HPR products are probably fine and may well be kinder to your barrier, but paying a premium for those claims is a bet rather than a sure thing. Strength, how often you use it, and what you pair it with all do more for your skin than which of these you pick off the shelf.

Where does adapalene fit for Asian skin?

Adapalene deserves its own answer, because it sits outside the ladder above and most beginner guides skip it entirely. It is a synthetic retinoid that binds retinoid receptors directly, so like tretinoin it needs no conversion. But unlike tretinoin it was designed for stability and selectivity, and it behaves very differently on the skin. In several intraindividual patch studies it came out as the least irritating of the common acne retinoids, gentler than the tretinoin and tazarotene gels it was tested against.

What makes it worth knowing about here is that the tolerability data actually includes darker skin, which is rarer than it should be. Callender and colleagues pooled three randomised vehicle-controlled trials covering 909 patients over 12 weeks, published in the Journal of Clinical and Aesthetic Dermatology (2010), and split the results by Fitzpatrick type. Dryness, scaling and stinging showed no significant difference between types I to III and types IV to VI. Redness actually went the other way: 59 percent of the darker-skinned group reported no erythema at all, against 45 percent of the lighter group, and fewer of them reported mild erythema (30 versus 40 percent, P less than 0.001). An earlier pooled analysis of single-agent adapalene 0.1% gel found the same pattern, with significantly less erythema and scaling in the darker-skinned subgroup.

That runs against the usual assumption, so hold it loosely. Reported irritation is not the same as measured barrier disruption, and redness is genuinely harder to see and to self-report on deeper skin tones. But the practical read still stands. Adapalene is not a harsher option you graduate into. For a lot of people it is the gentler starting point, and in much of Asia you can buy 0.1% over the counter without a prescription.

There is now Asian anti-ageing data too, not just acne data. A randomised controlled trial in Vietnam, published in the Indian Journal of Dermatology, Venereology and Leprology (2025), followed 58 women with a mean age of 42, evenly split between Fitzpatrick types III and IV, for six months. One group used adapalene 0.1% cream on top of a basic routine of cleanser, moisturiser and sunscreen. The other group used the routine alone. Total skin ageing scores fell from 38.2 to 32.5 in the adapalene group against 38.5 to 37.9 in the control group, with significant gains in both pigmentation and wrinkles.

The side effect numbers are the honest part, and they are not trivial. Stinging affected 64.3 percent of the adapalene group and dryness 25 percent, though both mostly settled after the first two months. About 10.7 percent got breakouts and the same proportion got hyperpigmentation, and those resolved by month two as well. So adapalene is gentler than tretinoin, not gentle in the absolute sense. The first eight weeks still ask something of you, which is the same lesson every section of this guide keeps landing on.

One practical note. Adapalene is famously stable, including alongside benzoyl peroxide, which most retinoids are not. If you are treating active breakouts as well as early ageing, that combination is a real advantage. If your concern is purely fine lines and tone, plain retinol or retinal remains the simpler place to begin.

Is retinol purging real, and how long does it last?

Purging is real, and it catches a lot of beginners off guard. Because retinol speeds up cell turnover, it can push along breakouts that were already brewing under the surface, so you get a cluster of spots in the first few weeks. That is purging, and it typically settles within about four to six weeks as your skin adjusts.

Here is how to tell purging from a bad reaction. Purging happens in the places you normally break out, and it fades. If you get new breakouts somewhere you never usually do, or persistent stinging, burning, and redness, that is irritation, not purging, and it is a sign to scale back. When in doubt, do less. There is no prize for pushing through pain, and on pigment-prone skin, pushing through can leave marks.

How do you avoid irritation?

Most retinol misery is avoidable with a few simple habits.

• Buffer with moisturiser. Try the sandwich method: moisturiser, then retinol, then moisturiser again. It softens the hit without cancelling the benefit.
• Never skip sunscreen. Retinol makes your skin more sun sensitive, and sun is the top trigger for pigmentation. Daily SPF is not optional here. Mar and colleagues reviewed 14 studies covering 369 patients with skin of colour in the Australasian Journal of Dermatology in 2025 and found sunscreen alone prevented post-inflammatory hyperpigmentation in 98.4% of cases, rising to 100% when paired with an anti-inflammatory. Nothing else in that review came close. Our sunscreen guide for Asian skin covers what to pick.
• Do not mix with other strong actives at first. Keep retinol away from exfoliating acids and vitamin C in the same routine while you are starting out, so your skin is not hit from all sides.
• If your skin gets angry, pause. Dryness, peeling, and persistent redness mean your barrier needs a break. Stop retinol for a week or two, baby your skin with a plain gentle moisturiser, then restart at a lower frequency.

Which soothing ingredients actually reduce retinol irritation?

This is where the advice usually goes vague, so here's a study that put numbers on it. Ying and colleagues ran a patch test on 21 healthy Chinese participants aged 21 to 35, applying 0.1% retinol on its own and then paired with different soothing agents, and published the results in the Journal of Cosmetic Dermatology in 2025. They scored each combination with a cumulative irritation index, where lower is calmer.

The pairings that performed best, none of which produced a single Grade 2 reaction:

• 2% PLG scored 0.26, the calmest of the lot
• 5% ceramides scored 0.42
• 3% acetyl glucosamine scored 0.51
• 2% panthenol scored 0.55
• 0.2% TECA scored 0.62
• 0.5% Centella asiatica scored 0.68

Look at that list again if you shop K-beauty or J-beauty. Ceramides, panthenol, centella and TECA are already the backbone of half the barrier creams on the shelf, so the thing that calms retinol is probably sitting in your routine already. Pairing them deliberately is the move. The researchers also saw a dose-dependent pattern, with 2% outperforming 0.8%, so concentration matters on the soothing side too.

Our guides to ceramides for Asian skin and centella asiatica go deeper on both.

A separate team found the same thing a year earlier, and independent replication is worth paying attention to. Fang and colleagues ran a five-day occlusive patch test on 18 Chinese participants aged 23 to 40, randomised, double-blind and vehicle-controlled, published in the Journal of Cosmetic Dermatology in 2024. They pitted 2% PLG against a 5% physiologic lipid mixture of ceramide 3 and cholesterol, and against 3% niacinamide. Both the PLG and the lipid mixture cut retinol-induced redness compared with the vehicle control, and PLG came out ahead again. A follow-up in-use test backed it up.

But one result cuts against what you'll read almost everywhere. The 3% niacinamide gave no relief from retinol irritation at all. That doesn't make niacinamide a bad ingredient, and it still earns its place alongside retinol for reasons we get into below. It just isn't the acute anti-irritant it gets marketed as. So if your skin is stinging tonight, niacinamide probably isn't the fix. Reach for the lipids instead.

Should you put retinol around your eyes?

Cautiously, and not right up to the lash line. Eyelid skin is the thinnest on the body, so the same amount that feels fine on your cheek can leave the eye area flaking for a week.

The evidence for retinoids specifically formulated for that zone is thinner than the marketing suggests. Kaufman, Callender and colleagues ran a 12-week dual-centre open-label trial of a retinoid eye cream and published it in the Journal of Drugs in Dermatology in 2022. In the 16 people using the eye cream on its own, fine lines improved 33 percent (P < .0001), under-eye darkness 41 percent (P < .001), and dryness 94 percent (P < .0001). No adverse events were judged related to the products.

Encouraging, but hold it loosely. Twenty-six people finished the study, there was no control group, and open-label means everyone knew what they were using. It tells you a well-formulated retinoid eye product can be tolerable. It doesn't tell you your face serum will be.

What works in practice is keeping a deliberate gap. Stay about a centimetre off the lash line and skip the mobile eyelid entirely. Apply your moisturiser to the eye area first so the retinol that migrates lands on a buffered surface. And if you want the orbital bone area treated, use a product made for it rather than dragging your facial strength around the socket.

Can you wash retinol off to reduce irritation?

Yes, and it is a genuinely underused trick. It is called short contact therapy, and it means applying your retinoid, leaving it on for somewhere between 30 and 60 minutes, then washing it off with a gentle cleanser. Your skin still absorbs a useful amount in that window. It just does not sit under the product all night while your barrier is trying to recover.

The clearest test of it is the MASCOTTE trial, run across two centres in Italy by Bertolani and colleagues and published in Dermatologic Therapy in 2021. Forty-six people with mild to moderate acne, mean age 23, were randomised to either standard overnight application or short contact with the same gel, which contained tretinoin 0.02%, clindamycin 0.8% and glycolic acid 4%. The short contact group washed the product off after one hour. Over eight weeks that group scored better on tolerability, measured across redness, dryness, stinging and burning, while efficacy held up at least as well as leaving it on.

Two caveats before you rearrange your routine around this. The trial used a prescription combination gel rather than an over-the-counter retinol serum, and it was small. Nobody has run the same comparison on the retinol you are likely to be holding. So treat this as a reasonable technique with supporting evidence rather than a proven protocol.

Where it earns its place is as a way in rather than a permanent setting. If you have tried retinol twice and both times ended up red and flaking, short contact gives you a third option that is not simply giving up. Start at 20 or 30 minutes, twice a week. Build the contact time up before you build the frequency up, and once an hour feels comfortable you can try leaving it on overnight. It is also useful in the other direction, as somewhere to retreat to during a stretch of travel, air-conditioning, or anything else that has left your barrier fragile.

What if you've already overdone it?

Stop the retinol. Not forever, just for now. Almost everyone's instinct is to push through and hope the skin adjusts, and that's the one move that reliably makes it worse.

You'll know you've crossed the line when the flaking stops looking like light peeling and starts looking raw. Stinging when you apply plain moisturiser is the clearest tell. So is skin that feels tight and hot, redness that doesn't settle overnight, or a sudden reaction to products you've used happily for years. That last one catches people out. If your gentle cleanser suddenly burns, your barrier is the problem, not the cleanser.

Here's the recovery plan, and the whole point is to do less rather than add more.

• Pause the retinol completely. Not a lower strength, not every third night. Off, until your skin stops complaining.
• Cut everything else back too. No acids, no vitamin C, no scrubs, no clay masks. A gentle cleanser, a moisturiser, and sunscreen. That's the whole routine for a week or two.
• Lean on the repair ingredients. The same ones that showed up best in Ying's patch test work here. Ceramides, panthenol, and centella are the ones to reach for, and a plain thick cream beats anything with a long active list.
• Be strict about sunscreen. Irritated skin pigments more easily, and this is where the dark marks come from. Skipping SPF during recovery is how a week of redness turns into three months of patchiness.
• Don't pick at the flaking. Peeling it off takes live skin with it and restarts the whole cycle.

Most people are back to baseline in one to two weeks. If you're still raw after three, or if you're seeing swelling, weeping, or spreading rash rather than plain dryness and redness, that's a dermatologist visit rather than a shopping problem. Contact dermatitis to something else in the formula is possible, and it needs a different fix.

When you do restart, go back in below where you were. Half the frequency, or a step down the retinoid ladder, and give it a fortnight before changing anything. The skin that just got irritated is not the skin that tolerated the old routine, and treating it like it is will land you straight back here. Our skin barrier repair guide covers the rebuild in more detail if you want the longer version.

And be honest with yourself about what caused it. Usually it isn't the retinol alone. It's the retinol plus an acid toner, or retinol on four nights when two was working, or a new strength started the week you also began an exfoliant. The fix is rarely a better product. It's fewer things at once.

Can you use retinol with vitamin C, acids, or niacinamide?

Some yes, some not yet. The blanket "never mix actives" rule you see everywhere is too crude, and it stops people using the one pairing that genuinely helps.

Niacinamide is the pairing worth making

Start here, because this is the useful one. There's an old internet claim that niacinamide and retinol cancel each other out or react badly. It doesn't hold up, and the evidence points the other way.

Fu and colleagues ran a randomised, parallel-group trial on 196 women aged 40 to 65, published in the British Journal of Dermatology in 2010. One group used a cosmetic regimen combining niacinamide, peptides and retinyl propionate. The other used prescription 0.02% tretinoin. At eight weeks the cosmetic regimen came out ahead, with 58 percent of users showing visible improvement against 41 percent on tretinoin. The tolerability gap was the striking part: tretinoin raised transepidermal water loss by 5.9 g/m2/h while the cosmetic regimen showed no change at all, and the tretinoin group had significantly more redness and dryness throughout.

Two honest caveats. By 24 weeks the 50 people who continued showed no meaningful difference between the two regimens, so tretinoin caught up. And the participants were Fitzpatrick skin types I to III, which is not the audience of this page, plus the study was industry funded. Hold the numbers loosely. The direction still matches the Chinese four-week study on retinol with niacinamide and ceramides. Worth being precise about what niacinamide is doing, though. Fang's patch test found it did nothing for acute retinol sting over five days, so it isn't a rescue ingredient for a bad night. What it does is support the barrier over weeks, and a better-supported barrier tolerates retinol better. Those are two different jobs, and niacinamide is only good at the second one. Our niacinamide guide covers how to slot it in.

Vitamin C, acids and benzoyl peroxide need spacing

• Vitamin C: use it in the morning and keep retinol for night. This is about your skin's tolerance rather than the ingredients destroying each other, which is the usual overstated version. Splitting them across the day gets you both without stacking the sting. Our vitamin C guide has the morning routine.
• Exfoliating acids (AHA, BHA): alternate nights rather than layering. Retinol already increases cell turnover, so adding acids on top is doubling up on the same job and multiplying the irritation. Once you're settled after a few months, once-weekly acid on a non-retinol night is plenty.
• Benzoyl peroxide: keep them apart, morning and night. Benzoyl peroxide can degrade some retinoids on contact, so applying them together may waste both.
• Hydrating and barrier ingredients: no spacing needed. Hyaluronic acid, ceramides, panthenol and centella can go on alongside retinol and generally make it easier, not harder.

The rule that matters more than any pairing chart: introduce one new active at a time and give it about four weeks. If you start three things in one week and your face reacts, you've learned nothing about which one did it. Our guide to layering skincare products covers the order.

Does your climate change how you use retinol?

It changes the sun half a lot, and the moisturiser half less than you'd expect. This is the section that actually differs from advice written for London or New York. But Asia is not one climate, so it splits in two. If you're in Singapore, KL, Jakarta, Manila or Bangkok, read the tropical part. If you're in Seoul, Tokyo, Taipei, Beijing or northern India, your hard season is the opposite one, and that part comes after.

The UV problem is bigger here, and it's year-round

Retinol makes skin more sun sensitive, and irritation plus UV is exactly the combination that leaves dark marks on pigment-prone skin. So the sunscreen instruction carries more weight near the equator than it does in a temperate country.

Singapore's National Environment Agency states that it's common for the UV Index to reach Very High and Extreme levels during the four hours between 11am and 3pm on days with little cloud cover. On the World Health Organization scale that means 8 to 10 for Very High and 11 or above for Extreme, which is the top of the chart. And because equatorial countries don't get a meaningful winter, that's a year-round condition rather than a summer one.

Practically, that means daily sunscreen isn't a nice-to-have while you're on retinol, and reapplication through the day matters more than it does in a milder climate. Shade and a hat during those midday hours do real work too. Our sunscreen reapplication guide covers how often, and the sunscreen guide for Asian skin covers what to pick.

Humidity helps, air-conditioning undoes it

Here's the part people get wrong. High humidity genuinely works in your favour on the dryness side, since ambient moisture slows water loss from the skin. Retinol flaking is often milder in a tropical climate than in a dry winter one, which is a real advantage.

But most of us in this region sleep and work in air-conditioning, and air-conditioned air is dry. So your skin spends the night in a dehydrating environment during exactly the hours retinol is doing its work. That's the hidden reason tropical users still get flaking despite the humidity outside.

A few adjustments that fit the climate:

• Don't drop the moisturiser, change its texture. A lighter gel-cream still buffers retinol without the heavy, clogged feeling of a rich balm in humid weather. Skipping moisturiser entirely because your skin feels oily is the common mistake here.
• Let your skin dry fully before applying. Applying retinol to skin that's still damp from a shower increases absorption and irritation. Wait ten to twenty minutes, which matters more when you're never quite dry.
• Watch the sweat. If you're heading out into heat soon after applying, you'll sweat much of it off and irritate what stays. Night application solves this, which is another reason the AAD's night-only rule fits the tropics well.
• Expect oily skin to still need the buffer. Oily and dehydrated at once is the standard tropical combination, and retinol pushes it further. Our guides to humid climate routines and oily skin in a tropical climate go deeper.

What if your winters are cold and dry instead?

Then flip the advice above, because your barrier is under most strain at exactly the opposite time of year. A lot of retinol writing aimed at Asia quietly assumes the equator, which leaves out Seoul, Tokyo, Taipei, most of China and the northern half of India, where winter is dry and heated indoor air makes it drier.

There's good measurement on this. Yang, Tu, Man and colleagues followed 72 women with a mean age of 23 through all four seasons in Kunming, China, measuring the same cheeks and forearms each time, and published it in Skin Research and Technology in 2020 (26(5), 702 to 707). Because it's self-controlled, the seasonal shifts are within the same people rather than between different groups. Stratum corneum hydration and sebum both peaked in summer and hit their lowest point in winter. Transepidermal water loss, which is the rate your skin leaks moisture, went the other way and peaked in winter.

So in a dry-winter city your skin is least hydrated and leakiest in the months when you're most likely to be starting a new active. That's a bad combination with retinol, and it's the reason so many people conclude retinol doesn't suit them after trying it for the first time in January.

What to do differently:

• Start in a kinder season if you can. Beginning in late spring or summer gives you a barrier that's measurably better hydrated while you're adapting. Not essential, but it stacks the odds.
• Go richer, not lighter. The gel-cream advice for the tropics is wrong for you in winter. A proper cream, and the sandwich method of moisturiser before and after retinol, does real work when the air is dry.
• Drop the frequency in the cold months. If you run three nights a week in summer, two may be all your skin takes in January. Going backwards for a season isn't failure.
• Sunscreen still isn't optional. Cold doesn't mean low UV, snow and high altitude both reflect it back at you, and retinol keeps you sun-sensitive year-round.

The general rule underneath both halves is the same. Retinol is harder on your skin whenever your barrier is already working hard, and which season that is depends entirely on where you live. Our skin barrier repair guide covers the repair side in more detail.

Does retinol go off, and how should you store it?

Yes, faster than almost anything else in your routine, and heat makes it dramatically worse. If you live somewhere tropical, this matters more for you than for the people writing most retinol advice.

Temova Rakusa and colleagues tested 16 retinoid derivatives across 12 commercial cosmetic products and published the results in the Journal of Cosmetic Dermatology in 2021, running six months of long-term and accelerated stability testing plus a one-week photostability study.

The numbers are worse than most people assume. After six months at 25 degrees Celsius, retinoid content fell by anywhere from 0 to 80 percent depending on the product. At 40 degrees, the decline ran from 40 percent all the way to 100 percent. Complete loss, in some cases, of the only ingredient you bought the product for.

Now think about where 40 degrees actually happens. A bathroom cabinet in Singapore after a hot shower. A car glovebox. A bag left by a window. A shelf above the sink in a flat with no air-conditioning. None of that is unusual, and none of it looks like mistreatment.

Two more findings worth knowing. Light degradation was more pronounced than the temperature effect, which means a clear jar on an open shelf is doing more damage than the warm room it sits in. And among everything tested, hydroxypinacolone retinoate held up best. That's the ingredient sold as granactive retinoid, so if you went for one of those, you bought stability as well as gentleness.

What to actually do about it:

• Keep it out of the bathroom if your bathroom gets steamy. A bedroom drawer is genuinely better.
• Opaque packaging and airless pumps beat jars. Every time you open a jar you let in light and oxygen, and you dip fingers in on top of that.
• Buy smaller sizes. A large bottle you use for a year is a bottle that's noticeably weaker by the end.
• Watch for the colour turning. Retinol drifting yellow or brown, or developing a sharp smell, means it's oxidised. That's the point to replace it.

There's a practical upside here. If a retinol that used to sting stops doing anything at all after several months, the honest explanation may be that the product faded rather than that your skin toughened up. Worth knowing before you jump to a higher strength you don't need. You can keep track of what you've opened and when with our product expiry tracker.

When will you see results?

Retinol is a slow burn, so set your expectations for the long game. The first few weeks are usually the awkward phase of purging and adjustment, not glow. Real, visible improvement in texture, tone, and marks tends to show up around the 8 to 12 week mark, and it keeps building the longer you stay consistent.

For fading dark marks specifically, retinol is a genuine help, since the same review work on skin of colour found topical retinoids to be among the most studied and effective treatments for post-inflammatory hyperpigmentation. But it works best as part of a full routine with sunscreen and a brightening ingredient or two, not as a lone hero. Give it three months of steady, gentle use before you judge it. Your future skin will thank you for not rushing.

Does retinol thin your skin?

No. It does close to the opposite. The living layers of your skin get thicker on retinol, and the deeper support structure gets stronger. The thinning worry is one of the most common reasons people quit, so it's worth seeing what the biopsies actually show.

The key experiment came out of the University of Michigan. Sewon Kang and colleagues applied retinol at up to 1.6 percent, tretinoin at 0.025 percent, or a plain base to normal skin under a covering for four days, then took small samples to look at under a microscope. Their results, in the Journal of Investigative Dermatology in 1995, showed retinol thickened the epidermis 1.5-fold, close to tretinoin's 1.6-fold. And here's the part beginners like. Retinol produced only trace redness that wasn't statistically different from the plain base, while tretinoin caused a 3.7-fold jump in redness in the 10 people tested.

So why does it feel thinner? Mostly because of what happens on the surface. In the first weeks the outermost layer of dead cells sheds faster, which shows up as flaking and a tight, sensitive feel. Your skin looks and feels more fragile for a while. But that's the top layer turning over, not the skin underneath wasting away. If it stings and peels, that's irritation to manage with the steps above, not a sign of damage building up.

What does retinol actually do to collagen?

It helps you make more of it, and one small trial tested this in very old skin, where you'd expect the least response. Reza Kafi, Sewon Kang and the Michigan team recruited 36 residents of two senior care homes, with an average age of 87. Each person had 0.4 percent retinol lotion put on one arm and a plain lotion on the other, up to three times a week for 24 weeks. The results appeared in Archives of Dermatology in 2007.

• Fine wrinkles improved. On a 0 to 9 scale, the retinol arm dropped by 1.64 points against 0.08 for the plain lotion.
• Collagen production went up. Staining for procollagen I, the building block of new collagen, rose on the retinol side in the subgroup that was biopsied.
• Water-holding molecules went up too. Glycosaminoglycans, which hold a lot of water in the skin, increased as well.

The authors went further than appearance. They wrote that with more of this matrix being built, retinol-treated aged skin "is more likely to withstand skin injury." That's the reverse of thinning.

A couple of honest limits. The biopsy findings came from small subgroups of 6 and 4 people, and neither study was done in Asian participants. But the biology here is basic skin structure rather than anything tied to skin tone, and the Asian-specific risk this guide keeps coming back to is irritation and dark marks, not thinning. If you're weighing a gentler option for that reason, our bakuchiol vs retinol comparison covers the trade-off.

Can you use retinol on your neck, chest or hands?

Yes, and the Kafi trial is a good reminder that body skin responds too, since it was done on arms. But neck and chest skin is thinner and has fewer oil glands than your face, so it gets irritated faster. Start at a lower frequency there than on your face, say twice a week, and use half the amount. Add a moisturiser on top if it stings. Hands get washed so often that retinol tends to come off quickly, so apply it at night and leave it on.

Sunscreen matters even more on these areas, because they're the parts people forget. Your neck, chest and the backs of your hands get a lot of everyday sun, especially in the tropics. Our sunscreen reapplication guide covers how to keep them protected through the day.

When should you pause retinol before a facial, laser, or waxing?

Days before, not hours. Retinoids speed up how fast your surface skin turns over, and anything that lifts or abrades that surface lands harder when retinol has been working on it. Facials, threading, waxing, peels and laser sessions are all common enough across Asia that this catches people out constantly.

Be warned that the evidence here is weaker than the confidence with which clinics state their rules. Balado-Simó and colleagues reviewed topical tretinoin across its dermatology uses in the Journal of Clinical Medicine in 2025 and were blunt about it. On peri-procedural timing they note the published protocols are mainly based on anecdotal evidence and personal experience rather than trial data.

The one concrete protocol they do report runs opposite to what most people assume. For ablative laser work, the approach described was 0.1% tretinoin for three months beforehand, stopped just one day before treatment. For non-ablative lasers, the strength was dropped to 0.05% instead. So retinoid prep can be part of the plan rather than something to purge from your routine, and your practitioner's instructions beat any general rule you read online.

That review also gives a useful baseline for how often retinoids bother people at all. In an open-label adult study of 544 participants, the adverse event rate was 13.6 percent and mostly mild. Pooled across 1,640 participants, the most frequent complaints were dryness at 5.0 percent and application site pain at 2.9 percent.

So what do you actually do?

• Ask first, and be specific. Tell whoever is treating you exactly what you use, the strength, and how often. A therapist who knows you're on 0.5% retinol four nights a week will adjust. One who assumes you use nothing will not.
• Stop roughly five to seven days out for anything that strips or pulls. Waxing and threading are the ones worth being strict about, because the failure mode is skin lifting off rather than mild redness. This is convention rather than trial-backed, which is a reason to be more careful, not less.
• Follow your clinic over the internet for lasers and peels. As the protocol above shows, some practitioners want you continuing retinoids right up to the procedure. Guessing from a blog post can undercut the result you paid for.
• Restart only when the skin is calm. Wait until any redness, peeling and tenderness have fully settled, then go back in at a lower frequency than before rather than straight to your old schedule.
• Don't stack a fresh retinol start with a booked treatment. If you have a facial or laser session coming up in the next fortnight, begin retinol after it, not before.

None of this is a reason to keep retinol and treatments apart forever. It's a scheduling problem, and scheduling problems are solvable.

What age should you start retinol?

There is no age that switches it on. Start when you have a reason to, not when you hit a birthday, and the reason is what should drive the decision.

People usually ask this because of ageing, so start with what actually causes it. Flament and colleagues assessed 298 women in Montpellier and published the results in Clinical, Cosmetic and Investigational Dermatology in 2013, attributing roughly 80 percent of visible facial ageing signs to sun exposure rather than to the passing of time. You can read it open access at Dove Medical Press. Worth naming the limit on that one: the sample was Caucasian women in southern France, so treat the 80 percent as the general shape of the thing rather than a number measured on Asian skin.

The useful part is what it implies. If most of what people call ageing is accumulated UV damage, then the clock that matters is exposure, not age. And as the tropical section above covers, you accumulate that faster near the equator, where the UV Index sits at Very High or Extreme through the middle of the day all year rather than for one season.

So what does that mean in practice?

• If you have acne, the trigger is the acne. Retinoids are an acne treatment first and an ageing treatment second. Plenty of people start in their teens or early twenties for that reason, usually with a dermatologist steering it.
• If you are seeing early texture or pigment changes, that is a reason. Not a number on a form. Uneven tone, marks that linger after a spot heals, fine lines that stay put when your face is still.
• If you have no concerns and just read that everyone should be on it, wait. Sunscreen is doing more for you at that point, and it costs you nothing in irritation.

That last one deserves saying plainly, because the internet pushes hard the other way. Starting early is not free. Retinol carries a real irritation cost, and on pigment-prone skin that irritation is the thing that leaves dark marks in the first place, which is the whole reason this guide exists. Taking on that risk for a problem you do not have yet is a bad trade.

If prevention is genuinely your goal and you have nothing to treat, daily sunscreen is the higher-yield habit by some distance. Our sunscreen guide for Asian skin covers what to pick, and if you want something gentler than retinol to start with, bakuchiol is the usual first step.

Which people should not use retinol?

Retinol suits most people, but there are a few clear exceptions and one of them is non-negotiable.

If you're pregnant or trying to conceive, skip it. The US Food and Drug Administration placed retinoids in pregnancy category X, its most restrictive class, meaning the known risk to a fetus outweighs any benefit. That classification came from the well-documented birth defects caused by oral retinoids like isotretinoin, and topical products absorb far less. But the American College of Obstetricians and Gynecologists and most dermatologists still advise avoiding all retinoids in pregnancy, including over-the-counter retinol and retinaldehyde. There are plenty of pregnancy-friendly alternatives, and bakuchiol is one worth looking at.

Beyond that, the AAD suggests talking to a dermatologist first if you have sensitive skin, rosacea, hormonal acne, or moderate to severe acne scarring, since a professional can pick a formula and pace that fits rather than leaving you to guess. And if your barrier is already in trouble, meaning tight, stinging, flaky skin that reacts to everything, fix that first. Retinol on a broken barrier is a fast route to exactly the irritation and dark marks you're trying to avoid. Our skin barrier repair guide covers how to get back to baseline before you start.

Can retinol help with melasma?

Sometimes, slowly, and with a real catch. Melasma is one of the most common pigment complaints across Asia, and retinoids do have evidence behind them. But almost all of that evidence is for prescription tretinoin rather than the retinol you buy over the counter, and the irritation that retinoids cause can make pigment worse before it makes it better.

Start with the strongest Asia-specific trial, because it's genuinely old and genuinely good. Griffiths and colleagues at the University of Michigan ran a double-blind randomised study on 45 photoaged patients, 23 Chinese and 22 Japanese, published in the Journal of the American Academy of Dermatology in 1994. Twenty-one applied 0.1% tretinoin cream daily, 24 applied a vehicle cream, and they kept it up for 40 weeks.

The results were clear. Hyperpigmented lesions were lighter or much lighter in 90 percent of the tretinoin group against 33 percent on vehicle. Biopsies showed a 41 percent drop in epidermal pigmentation with tretinoin, while the vehicle group went the other way with a 37 percent increase. Nobody withdrew because of side effects.

Now read the timeline again. Forty weeks. That's not a typo, and it's the single most useful number in this section. Retinoids move pigment on a scale of many months, not the four to six weeks people usually give a product before deciding it doesn't work.

A 2009 evidence review in American Journal of Clinical Dermatology, led by Kang at Ajou University School of Medicine in Suwon, graded the literature formally (PubMed 19489658). Topical tretinoin as a standalone melasma treatment got grade B, meaning fair evidence. The fixed triple combination of hydroquinone 4%, tretinoin 0.05% and fluocinolone acetonide 0.01% also got grade B. So retinoids help, but they're rarely the whole answer.

That triple combination was tested head to head in Asian skin specifically. A multicentre randomised trial across East and South-East Asia, published in the British Journal of Dermatology in 2008 and led from the National Skin Centre, compared it against hydroquinone 4% alone over eight weeks (PubMed 18616780). At week eight, 64.2 percent on the triple combination had melasma rated none or mild, against 39.4 percent on hydroquinone alone. Both of those are prescription routes, and both need a doctor.

Does over-the-counter retinol do the same thing?

Not at the same speed, and probably not to the same degree. As covered further up, retinol has to convert in the skin before it does anything, and the commonly cited potency gap against tretinoin is roughly ten to one. So a 0.3% retinol is not standing in for 0.1% tretinoin, and none of the melasma trials above tested it.

What you can reasonably expect from retinol is gradual help with the general dullness and uneven tone that sits alongside melasma, over a long stretch. What you should not expect is the patch itself clearing on its own.

And here's the catch that matters most for Asian skin. Melasma sits on skin that pigments easily when it's inflamed. Push the retinol too hard, get red and flaky, and you can trigger post-inflammatory hyperpigmentation that lands right on top of the problem you were trying to fix. Every go-slow instruction in this guide applies double if melasma is your reason for starting.

Two things are non-negotiable if you go down this road. Daily sunscreen, reapplied, because melasma is driven by UV and visible light and no topical will outrun an unprotected face. And patience measured in seasons rather than weeks. Melasma also relapses, so treat it as something you manage rather than something you finish. Our guide to hyperpigmentation on Asian skin covers the wider set of options, and the sunscreen guide covers the protection half properly.

Which studies is this based on?

Kang, S., Duell, E.A., Fisher, G.J. et al. (1995), "Application of retinol to human skin in vivo induces epidermal hyperplasia and cellular retinoid binding proteins characteristic of retinoic acid but without measurable retinoic acid levels or irritation," Journal of Investigative Dermatology 105(4):549 to 556 (PubMed).  ·  Kafi, R., Kwak, H.S., Schumacher, W.E. et al. (2007), "Improvement of naturally aged skin with vitamin A (retinol)," Archives of Dermatology 143(5):606 to 612, 36 older adults, mean age 87 (PubMed).  ·  Flament, F., Bazin, R., Rubert, V., Simonpietri, E., Piot, B. and Laquieze, S. (2013), "Effect of the sun on visible clinical signs of aging in Caucasian skin," Clinical, Cosmetic and Investigational Dermatology 6:221 to 232, 298 women in Montpellier, Caucasian sample (Dove Medical Press).  ·  Callender, V.D., Baldwin, H., Cook-Bolden, F.E., Alexis, A.F., Stein Gold, L. and Guenin, E. (2022), "Effects of Topical Retinoids on Acne and Post-inflammatory Hyperpigmentation in Patients with Skin of Color: A Clinical Review and Implications for Practice," American Journal of Clinical Dermatology 23(1):69 to 81 (PubMed 34751927). Bluemke, A. et al. (2022), "Multidirectional activity of bakuchiol against cellular mechanisms of facial ageing," International Journal of Cosmetic Science (PMC9328396), industry funded. "The Anti-Acne and Reduction of Hyperpigmentation Effects of Products Containing Retinol, Niacinamide, Ceramides and Dipotassium Glycyrrhizinate in Chinese Women," Cosmetics 2025 (mdpi.com). Mar, K. et al. (2024), "Treatment of Post-Inflammatory Hyperpigmentation in Skin of Colour: A Systematic Review," Journal of Cutaneous Medicine and Surgery. Griffiths, C.E., Kang, S., Ellis, C.N. et al. (1995), "Two concentrations of topical tretinoin cause similar improvement of photoaging but different degrees of irritation," Archives of Dermatology 131(9), 99 patients over 48 weeks (PubMed 7544967). Mar, K. et al. (2025), "Prevention of Post-Inflammatory Hyperpigmentation in Skin of Colour: A Systematic Review," Australasian Journal of Dermatology, 14 studies, 369 patients (PMC12062726). Chang, J. et al. (2024), "Racial differences in tolerability of topical retinoids: A 15-year single-center retrospective cohort study," JAAD International vol. 16, 753 matched patients (PMC11217678). Ying, Y., Fang, Y., Chenlan, X. et al. (2025), "Evaluating the Efficacy of Soothing Agents in Mitigating 0.1% Retinol-Induced Skin Irritation: A Patch Test," Journal of Cosmetic Dermatology, 21 Chinese participants (PMC12494960). Fang, Y. et al. (2024), "Mitigation of retinol-induced skin irritation by physiologic lipids: Evidence from patch testing," Journal of Cosmetic Dermatology, five-day occlusive patch test, 18 Chinese participants aged 23 to 40, randomised, double-blind, vehicle-controlled (PubMed 38628085). Zasada, M. et al. (2020), "A Clinical Anti-Ageing Comparative Study of 0.3 and 0.5% Retinol Serums: A Clinically Controlled Trial," Skin Pharmacology and Physiology 33(2):102 to 116, split-face, 37 women over 12 weeks (Karger). American Academy of Dermatology Association, "Retinoid or retinol?" public guidance on starting frequency, formula strength and sun protection (aad.org). "A Randomized, Double-blind, Split-face Study Comparing the Efficacy and Tolerability of Three Retinol-based Products vs. Three Tretinoin-based Products in Subjects With Moderate to Severe Facial Photodamage," Journal of Drugs in Dermatology, source of the roughly ten-to-one retinol to tretinoin potency ratio (JDD). Fu, J.J.J., Hillebrand, G.G., Raleigh, P. et al. (2010), "A randomized, controlled comparative study of the wrinkle reduction benefits of a cosmetic niacinamide/peptide/retinyl propionate product regimen vs. a prescription 0.02% tretinoin product regimen," British Journal of Dermatology 162(3):647 to 654, 196 women, Fitzpatrick I to III, industry funded (PMC2841824). Yang, J., Tu, Y., Man, M.Q., Zhang, Y., Cha, Y., Fan, X., Wang, Z., Zeng, Z. and He, L. (2020), "Seasonal variations of epidermal biophysical properties in Kunming, China: A self-controlled cohort study," Skin Research and Technology 26(5):702 to 707. 72 women, mean age 23, all four seasons, same cheek and forearm sites (PubMed 32196767). National Environment Agency Singapore, "UV Radiation and UV Index," on Very High and Extreme UVI between 11am and 3pm (nea.gov.sg). Fluhr, J.W., Vienne, M.P., Lauze, C., Dupuy, P., Gehring, W. and Gloor, M. (1999), "Tolerance Profile of Retinol, Retinaldehyde and Retinoic Acid under Maximized and Long-Term Clinical Conditions," Dermatology 199(Suppl 1):57 to 60, 14-day patch test plus 44-week study in 355 participants (Karger). Ye, Y., Li, X., Xu, Y. and Wei, X. (2023), "Improvement of mild photoaged facial skin in middle-aged Chinese females by a supramolecular retinol plus acetyl hexapeptide-1 containing essence," Skin Health and Disease 3(4):e239, 32 women over 8 weeks, single-arm (PMC10395635). Shen, et al. (2026), "An Innovative Serum With Retinol, Hydroxypinacolone Retinoate, Peptides, and Silybin Improves Mild Photoaged Facial Skin in Middle-Aged Chinese Women," Journal of Cosmetic Dermatology 25(1):e70627, 31 women over 8 weeks, single-arm, proprietary concentrations (PMC12741608). Deda, A., Odrzywołek, W., Lebiedowska, A. et al. (2026), "Comparative Evaluation of Topical Stabilized Retinaldehyde 0.1% vs. 0.05% on Skin Biophysical and Biomechanical Parameters," Skin Research and Technology 32(2):e70326, split-face, 56 women over 24 weeks (PMC12928007).  ·  Temova Rakuša, Ž., Škufca, P., Kristl, A. and Roskar, R. (2021), "Retinoid stability and degradation kinetics in commercial cosmetic products", Journal of Cosmetic Dermatology, 20(7), 2350 to 2358. 16 retinoid derivatives across 12 products, six-month stability testing.  ·  Balado-Simó et al. (2025), "An Updated Review of Topical Tretinoin in Dermatology: From Acne and Photoaging to Skin Cancer", Journal of Clinical Medicine. Peri-procedural protocols described by the authors as based mainly on anecdotal evidence (PMC12653878).  ·  Kaufman, J., Callender, V.D., Young, C., Jones, P., Wortzman, M. and Nelson, D.B. (2022), "Efficacy and Tolerability of a Retinoid Eye Cream for Fine to Moderate Wrinkles of the Periorbital Region", Journal of Drugs in Dermatology. Dual-centre open-label trial, 26 subjects completed, 12 weeks (JDD).  ·  Callender, V.D., Preston, N., Osborn, C., Johnson, L. and Gottschalk, R.W. (2010), "A Meta-analysis to Investigate the Relation Between Fitzpatrick Skin Types and Tolerability of Adapalene-Benzoyl Peroxide Topical Gel in Subjects with Mild or Moderate Acne", Journal of Clinical and Aesthetic Dermatology. Three randomised vehicle-controlled trials pooled, 909 patients over 12 weeks (JCAD).  ·  "Effectiveness and tolerability of adapalene cream 0.1% in the treatment of female skin ageing: A randomised controlled trial" (2025), Indian Journal of Dermatology, Venereology and Leprology. 58 women in Vietnam, mean age 42, Fitzpatrick III and IV, six months (IJDVL).  ·  Bertolani, M. et al. (2021), "Efficacy and tolerability of short contact therapy with tretinoin, clindamycin, and glycolic acid gel in acne: The MASCOTTE study", Dermatologic Therapy 34:e14724. Randomised, assessor-blinded, two centres in Italy, 46 patients over 8 weeks (PubMed 33377285).  ·  Griffiths, C.E., Goldfarb, M.T., Finkel, L.J. et al. (1994), "Topical tretinoin (retinoic acid) treatment of hyperpigmented lesions associated with photoaging in Chinese and Japanese patients: a vehicle-controlled trial," Journal of the American Academy of Dermatology 30(1):76 to 84. 45 patients (23 Chinese, 22 Japanese), double-blind randomised, 40 weeks (PubMed 8277035).  ·  Kang, H.Y., Valerio, L., Bahadoran, P. and Ortonne, J.P. (2009), "The role of topical retinoids in the treatment of pigmentary disorders: an evidence-based review," American Journal of Clinical Dermatology 10(4):251 to 260. Grade B for tretinoin monotherapy in melasma (PubMed 19489658).  ·  Chan, R., Park, K.C., Lee, M.H. et al. (2008), "A randomized controlled trial of the efficacy and safety of a fixed triple combination (fluocinolone acetonide 0.01%, hydroquinone 4%, tretinoin 0.05%) compared with hydroquinone 4% cream in Asian patients with moderate to severe melasma," British Journal of Dermatology 159(3):697 to 703. Multicentre, East and South-East Asian patients, 8 weeks (PubMed 18616780).

What else do people ask about retinol?

Starting a new active can stir up more anxiety than it should, and if you tend to overthink these things, the free psychology quizzes at Decode Within are a fun place to learn about yourself.

Which guides should you read next?