Prescription retinoids have the trial evidence. Retinol has a handful of tiny studies

Topical retinoids are the one category of anti-ageing skincare with a real randomised evidence base, and prescription tretinoin is where nearly all of it sits. But the effect sizes in those trials are more modest than the category's reputation, the vehicle creams they were tested against performed surprisingly well, and the handful of studies comparing prescription tretinoin with over-the-counter retinol are far too small to settle the question most shoppers are actually asking.

The photoaging evidence

A systematic review restricted to randomised controlled trials of topical tretinoin for photoaging screened 180 studies and found seven that qualified [s1]. Seven is the entire randomised literature for the flagship anti-ageing ingredient over a twenty-year search window.

Within those seven, the direction of effect was consistent. Concentrations ranged from 0.025% to 5%, treatment durations from three months to 24 months, and every trial reported improvement in the clinical appearance of photoaging — wrinkling, mottled hyperpigmentation, sallowness and lentigines — with changes appearing as early as one month and persisting at 24 months [s1]. All of the studies reported tretinoin to be safe and well tolerated [s1]. The largest was a multicentre double-blind trial of tretinoin emollient cream 0.05% against a vehicle emollient cream in 205 participants over 24 months, in which 45 participants dropped out [s1].

The review's own stated limitation is the field's real problem: different tretinoin formulations, different outcome measures, and few recent studies, leaving no uniformity across the evidence [s1]. Four of the seven trials enrolled only women, and women were the majority in the rest [s1].

What the vehicle does

The acne literature is larger and better controlled, and it exposes something the marketing never mentions: a substantial fraction of the improvement in a retinoid trial happens in the arm that gets no retinoid.

In a post hoc analysis of randomised trials in 766 Hispanic participants aged 11 to 50 with moderate-to-severe acne, tretinoin 0.05% lotion reduced inflammatory lesions by 60.1% against 51.1% for the vehicle, and non-inflammatory lesions by 53% against 38.7% [s2]. Treatment success — a two-grade improvement on the global severity scale — was reached by 19.6% on tretinoin against 12.7% on vehicle [s2]. The drug wins, and it wins by roughly nine percentage points on inflammatory lesions and seven on treatment success.

The pattern repeats. In a double-blind multicentre trial of 178 participants, tretinoin microsphere 0.04% gel reduced lesion counts by 35.5%, 38.2% and 33.6% across three measures against 20.9%, 19.2% and 20.4% for the vehicle [s2]. In 110 preadolescents aged 9 to 11, tretinoin microsphere 0.04% reduced non-inflammatory lesions by 19.9 against 9.7 for the vehicle, and produced a two-point global assessment improvement in 25.5% against 13.0% — but showed no significant difference from vehicle on static global assessment or on inflammatory lesions at all [s2].

The American Academy of Dermatology's acne guideline gives topical retinoids a strong recommendation, alongside benzoyl peroxide, topical antibiotics and oral doxycycline [s3]. That grading reflects consistency of evidence across many trials, not the size of the effect in any one of them.

Prescription against over-the-counter

Three comparisons exist, and all are small.

A split-face study in 48 patients with photodamage compared tretinoin 0.025% cream with a double-conjugated retinoid cream over 12 weeks: similar improvement in both, with tretinoin associated with more irritation [s2]. A mechanistic study in 24 patients with moderate-to-severe photoaging compared tretinoin 0.02% cream with retinol plus retinyl esters over 24 weeks: similar clinical improvement, with tretinoin increasing erythema [s2]. Against that, a 2024 systematic review comparing tretinoin at concentrations from 0.02% to 0.1% with retinol, glycolic acid and other agents concluded that tretinoin was superior in efficacy, with irritation and erythema more frequent at higher concentrations [s2].

Two trials of 48 and 24 people finding no difference do not establish equivalence — they are underpowered to detect one. A systematic review finding superiority across heterogeneous comparators does not quantify by how much. The honest summary is that the difference between prescription tretinoin and cosmetic retinol has not been measured well enough to state, and that what has been measured suggests the trade-off runs through tolerability as much as through potency.

There is also a middle category that gets overlooked. Adapalene, a synthetic retinoid, is sold both on prescription and over the counter in the United States, where Differin gel and several generic adapalene gels carry over-the-counter marketing status while the cream formulations and the benzoyl peroxide combinations remain prescription-only [s4]. In a randomised trial of 86 women with facial photoaging, adapalene 0.3% gel and tretinoin 0.05% cream over 24 weeks produced comparable improvement in periorbital and forehead wrinkles, pigmentation and overall photodamage scores, with similar tolerability and no significant difference in efficacy [s2].

What retinoids do not do

A large randomised trial found no preventive effect of tretinoin on keratinocyte carcinomas [s2]. The review that reports it is otherwise expansive about tretinoin's off-label uses — melasma, post-inflammatory hyperpigmentation, striae, flat warts, alopecia areata, androgenetic alopecia, hypertrophic scars and actinic keratosis — but describes the supporting evidence for all of them as smaller randomised and prospective studies, and states that adequately powered trials with standardised outcome measures are still needed before clinical guidelines can be written [s2].

On safety, the same review's position is that adverse events are typically mild, localised and transient, and that available evidence does not support an association with systemic adverse effects [s2].

This article is informational and is not medical advice. Decisions about prescription retinoids belong with a clinician, and retinoid use in pregnancy carries specific restrictions not covered here.

Sources

• [s1] Topical tretinoin for treating photoaging: A systematic review of randomized controlled trials — International Journal of Women's Dermatology, 25 March 2022. https://doi.org/10.1097/jw9.0000000000000003
• [s2] An Updated Review of Topical Tretinoin in Dermatology: From Acne and Photoaging to Skin Cancer — Journal of Clinical Medicine, 10 November 2025. https://doi.org/10.3390/jcm14227958
• [s3] Guidelines of care for the management of acne vulgaris — Journal of the American Academy of Dermatology, 30 January 2024. https://doi.org/10.1016/j.jaad.2023.12.017
• [s4] Drugs@FDA: adapalene product listings and marketing status — US Food and Drug Administration, openFDA drug/drugsfda endpoint. https://api.fda.gov/drug/drugsfda.json?search=openfda.generic_name:%22adapalene%22