The Retinoid Evidence Base: What the Research Actually Says (Versus What the Internet Sells You)

If one skincare ingredient category has the most robust evidence and the most confusing internet landscape, retinoids would win every time.

The evidence base for some retinoids goes back to the 1980s. The number of peer-reviewed studies is substantial and the clinical outcomes are measurable. Yet the average woman encountering "retinol" content online is served a mixture of brand recommendations, vague reassurances that "retinol works", and very little explanation of why the ingredient she is buying is not the same one tested in the studies she just read about.

This post offers something more useful: a clear account of what the evidence actually shows, where the gaps are, and what it means in practice.

The retinoid hierarchy: why not all vitamin A is the same

Retinoids (the umbrella term for vitamin A derivatives used in skincare and medicine) are not a single ingredient. They are a family, and the differences between them matter clinically.

Think of the family as a conversion ladder. Every form of retinoid has to be converted in the skin to retinoic acid (the active form that binds to receptors in skin cells and produces clinical effects) before it can do anything. The higher up the ladder an ingredient starts, the fewer conversion steps it needs, and the more direct its action.

The hierarchy, from most to least potent:

• Tretinoin is retinoic acid. It sits at the top because it requires no conversion at all. It is the active form, and it is prescription-only in the UK.
• Retinaldehyde (also called retinal) sits one step below. It needs one conversion step to become retinoic acid and is available without a prescription, typically between 0.05% and 0.1%.
• Retinol needs two conversion steps: first to retinaldehyde, then to retinoic acid. It is the most widely available over-the-counter (OTC, meaning available without a prescription) retinoid in the UK.
• Retinol esters, including retinyl palmitate and retinyl acetate, need three conversion steps and are the mildest form, found in many basic moisturisers and eye creams.

The conversion process is not lossless. With each step, some of the starting material is lost. This is why a 1% retinol product does not produce the same skin response as a 0.025% tretinoin, even though the retinol number sounds higher. They are not equivalent.

What the evidence actually shows

Tretinoin: the gold standard with four decades of data

The evidence base for tretinoin is the oldest and most extensive of any topical retinoid. The first clinical trials in photoageing (skin damage caused by UV exposure over time) were published in the late 1980s by Kligman and colleagues, and the findings have been replicated many times since.

Research shows consistently that tretinoin, at concentrations between 0.025% and 0.1%, increases skin cell turnover, supports collagen production in the dermis (the deeper layer of skin beneath the surface), reduces fine lines, improves texture and tone, and fades pigmentation. These are histologically confirmed outcomes, meaning they have been verified by looking at skin biopsies before and after treatment. A 1995 randomised controlled trial (RCT, where participants are randomly allocated to treatment or placebo to reduce bias) in the Archives of Dermatology found that 0.05% tretinoin over 48 weeks produced significant reductions in fine wrinkles, roughness, and hyperpigmentation. The effects were dose-dependent: the 0.1% group improved more than the 0.025% group, but also had more irritation.

Retinaldehyde: promising, but the data is thinner

Retinaldehyde is the closest OTC option to tretinoin in terms of conversion steps, and earlier studies suggested it was well tolerated. A 1999 study by Creidi and colleagues compared retinaldehyde 0.05% to retinol 0.05% over 18 weeks and found retinaldehyde produced greater improvements in fine wrinkles and texture. These are interesting findings, but a limited evidence base: far fewer RCTs than tretinoin, generally shorter studies, and inconsistent concentrations. Retinaldehyde is genuinely promising, but calling it equivalent to tretinoin overall goes further than current research supports. Clinically, it is a reasonable option for those who want something stronger than retinol without a prescription, ideally at 0.05% or above.

For readers ready to start or step up a retinoid, these are well-formulated options across the strength range:

Medik8 Crystal Retinal 1

£39.50

SkinBetter AlphaRet Overnight Cream

£70

AllSkin Med R Advanced Renewal Serum

£90

NEOSTRATA Correct Comprehensive Retinol 0.3% Night Serum

£64

Retinol: well tolerated, but dependent on conversion

Retinol is the most widely studied OTC retinoid. There is good evidence that it improves texture, reduces the appearance of fine lines, and supports collagen production over time, because it eventually converts to retinoic acid in the skin. The limitation is the conversion itself: because retinol undergoes two steps before becoming active, a meaningful proportion of what you apply never reaches the active form, and the amount that converts varies by individual skin chemistry, concentration, formulation, and pH. That variability is one reason results with retinol are more inconsistent between individuals than with tretinoin. What the research confirms is that retinol produces measurable outcomes when used consistently over time (think twelve weeks or more at an adequate concentration) and is considerably better tolerated than tretinoin during the initial adjustment period. For most women starting retinoids, retinol at 0.3% to 1% is a sensible, evidence-supported entry point.

The common mistake is stopping too early. The first four to six weeks often bring some dryness, flaking, or redness. This is the retinoid reaction, it is temporary, and it usually means the concentration needs to be introduced more slowly, not abandoned.

Adapalene: the retinoid that changed categories

Adapalene sits slightly apart from the rest of the family. It is a third-generation synthetic retinoid, designed to bind to certain retinoid receptors rather than all of them, which is part of why its side effect profile differs from tretinoin. It was originally a prescription acne treatment, and research shows it is highly effective at reducing inflammatory acne lesions and preventing new ones. In 2016, lower-concentration adapalene (0.1%) became available over the counter in the UK, making it the most accessible clinically validated retinoid for acne-prone skin. Studies comparing it to tretinoin for acne generally find similar efficacy with better tolerability. A 2014 study in the Journal of Drugs in Dermatology found adapalene 0.3% (prescription strength) produced significant improvements in fine lines, texture, and hyperpigmentation over 24 weeks, comparable in several measures to tretinoin 0.025%. For women managing both breakouts and early signs of ageing, that is a useful return on one ingredient.

Related reading: Retinoids are powerful, but a healthy barrier comes first. If your skin reacts easily, read The Gut-Skin Axis: How Your Microbiome Impacts Eczema and Rosacea.

The UK prescribing context

In the UK, tretinoin requires a prescription. It is not available over the counter and cannot be legally sold without one. A significant portion of "retinol vs tretinoin" content online comes from regions where access is different, and the UK context gets overlooked. If you have used OTC retinol for some time without the results you hoped for, tretinoin may be more appropriate, and that is a conversation to have with a prescribing clinician who understands skin: a dermatologist, a GP with an interest in aesthetics, or an aesthetics practitioner with independent prescribing capability. The best retinoid for your skin is not always the one you can pick up on the high street.

What the evidence has clarified over time

Earlier practice often defaulted heavily to retinol for almost everyone, largely because it was accessible and seemed safe to recommend without detailed assessment. What the evidence has clarified is that "accessible" and "most appropriate" are not the same thing. For many people, starting with retinol is correct: effective enough, better tolerated, and gentle on the adjustment period. For others, particularly those with significant photoageing, established hyperpigmentation, or persistent acne, starting at the lower end of the ladder can mean months of sub-optimal results when a more targeted approach would serve them better.

The evidence has also strengthened the case for retinaldehyde as a genuine middle step for those who found retinol too irritating, and it has changed thinking on frequency. Early guidance often suggested nightly use from the start. We now know that starting two to three nights per week, building up over four to six weeks, and treating it as a skin adaptation process rather than a product swap reduces dropout and produces better long-term outcomes.

Common mistakes worth avoiding

• Starting too strong. With tretinoin especially, beginning at 0.05% or 0.1% when 0.025% was the right entry point leads to severe irritation and abandonment of an otherwise excellent treatment.
• Using a retinoid on a compromised barrier. If skin is already red, dry, or reactive, a retinoid makes it worse. Sort the cleanser and moisturiser first.
• Layering incorrectly. Retinoids go on after cleansing and before moisturiser on dry skin. Applying to damp skin or layering directly over acids or vitamin C increases irritation without increasing benefit.
• Inconsistent use. Three nights one week, none the next, is how people use retinoids for six months and conclude they do not work.
• Skipping SPF. Retinoids increase photosensitivity. Broad-spectrum SPF 50 every morning is non-negotiable, whatever the weather.

What this means in practice

If you are new to retinoids and your skin is generally healthy, retinol at 0.3% to 0.5%, two to three nights per week, is a reasonable start. Give it twelve weeks. If you have oily, acne-prone skin, adapalene 0.1% OTC is well evidenced and may address breakouts and texture at once. If you have used retinol for over six months with consistent use and limited results, retinaldehyde or a prescription option may suit your skin better. And if you have established photoageing, significant pigmentation, or acne scarring, tretinoin, on prescription, has the strongest evidence base for those concerns. For very sensitive skin or pregnancy, a bakuchiol-based product is a gentler alternative to discuss with your clinician.

Shop the products in this post

The right retinoid depends on where your skin is now, at what strength, and used consistently. These are well-formulated options across the range, plus the barrier and SPF support that make retinoids work:

Medik8 Crystal Retinal 1

£39.50

SkinBetter AlphaRet Clearing Serum

£134

AllSkin Med R Advanced Renewal Serum

£90

NEOSTRATA Correct Comprehensive Retinol 0.3% Night Serum

£64

SkinBetter Intensive AlphaRet Overnight Cream

£134

AllSkin Med Barrier Recovery Cream

£84

PRIORI Tetra fx251 Broad Spectrum SPF 50 Tinted Sunscreen

£87

Browse the full range at deboratentis.com. To find the right retinoid for your skin, book a Happy Skin Holistic Consultation at Debora Tentis Clinic (£60).

Keep reading: Oestrogen has a direct effect on collagen and skin quality. For how hormones shape your skin, read Oestrogen and Skin: The Clinical Connection.

Debora Tentis is a Women's Health Pharmacist and Independent Prescriber Trainee at Debora Tentis Clinic, Milton Keynes. This post is for educational purposes only and does not constitute medical advice. Retinoid selection depends on individual skin type, health history, and clinical assessment.