In 36 elderly adults (mean age 87 years), Kafi and colleagues reported in Archives of Dermatology in 2007 that 0.4% retinol (a vitamin A ingredient) lotion applied to one arm up to 3 times a week for 24 weeks improved fine wrinkling scores significantly more than the same lotion without retinol on the other arm. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A reading of the Kafi paper, or a repeat vehicle-controlled trial in elderly adults, that finds no significant difference in fine wrinkling change between 0.4% retinol lotion and the same lotion without retinol at 24 weeks would show this wrong.
In a systematic review of twelve randomized trials of topical benzoyl peroxide (an acne medicine), Lamel and colleagues reported in Archives of Dermatological Research in 2015 that total acne lesions fell by an average of 44.3 percent in benzoyl peroxide groups (2818 patients) and 27.8 percent in vehicle groups (2004 patients given the base without the medicine). source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A recount of the same twelve trials showing an average total lesion reduction in vehicle groups near zero, or no smaller than in the benzoyl peroxide groups, would show this wrong.
The U.S. Food and Drug Administration says the law does not require cosmetics to be FDA approved before sale, while drugs need FDA approval for both safety and effectiveness first. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A statute or FDA rule requiring cosmetics to receive FDA approval before they go on the market would show this wrong.
Griffiths and colleagues reported in Archives of Dermatology that tretinoin (a prescription retinoid) at 0.1% and 0.025% produced similar changes in photoaging (sun damage) despite significantly more irritation at 0.1%, which they say suggests mechanisms other than irritation dominate the repair. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A controlled trial finding that tretinoin's photoaging improvement rises and falls with the amount of irritation it causes would show this wrong.
Griffiths and colleagues reported in Archives of Dermatology that tretinoin (a prescription retinoid) at 0.1% and 0.025% produced similar clinical and histologic changes (visible changes and changes seen under the microscope) in patients with photoaging (sun damage), despite significantly greater incidence of irritation with the higher concentration. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A trial in photoaged patients finding that 0.1% tretinoin produced significantly different changes from 0.025%, or caused no greater incidence of irritation, would show this wrong.
In pig skin tests reported in Dermatologic Surgery, Pinnell and colleagues found L-ascorbic acid (pure vitamin C) had to be formulated below pH 3.5 (quite acidic) to enter the skin, and 20% was the maximal concentration for optimal absorption. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A skin delivery study finding that L-ascorbic acid enters pig skin from a formula at pH 3.5 or higher, or that absorption keeps rising above 20%, would show this wrong.
In tests on hairless guinea pig skin and human skin reported in Pharmaceutical Research, Boehnlein and colleagues found human skin absorbed about 18% of retinyl palmitate (a vitamin A ester) applied in acetone, and about 44% of that absorbed share was converted into retinol. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A skin absorption study finding that human skin does not convert absorbed retinyl palmitate to retinol, or converts a share far from 44% under the same conditions, would show this wrong.
In lab tests on human skin samples reported in Toxicology and Applied Pharmacology, Yourick and colleagues applied 0.3% retinol in two bases and found 1.3% of the applied dose passed through skin in 24 h from an oil-in-water emulsion (a lotion type), versus 0.3% from a hydroalcoholic gel (an alcohol and water gel). source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A human skin absorption test of the same 0.3% retinol finding similar amounts passing through skin in 24 h from the gel and the oil-in-water emulsion would show this wrong.
In vitamin C (ascorbic acid) cream formulations tested by Ahmad and colleagues in AAPS PharmSciTech, the vitamin still broke down by oxidation in the dark, but about seventy times slower than under the UV light used in the study. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A study of ascorbic acid creams finding that dark storage slows breakdown far less than about seventy times compared with UV light exposure would show this wrong.
In lab tests on skin cells reported in Toxicology Letters, Kim and colleagues found retinoid-induced inflammation was mainly mediated through two inflammatory messengers, MCP-1 and IL-8, whose gene activity and protein release rose in human skin cells treated with 10 microM retinol or retinoic acid. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A study finding that retinol-induced inflammation proceeds just as strongly when MCP-1 and IL-8 release is blocked would show this wrong.
A single-blind International Journal of Molecular Sciences study reported that 0.104% and 0.247% retinol creams, used 12 weeks by 28 randomized women aged 40-61 (Fitzpatrick phototypes II-III, lighter skin types), both significantly improved hydration and TEWL (skin water loss) with no inter-group differences. Consumer assessment indicated comparable efficacy; 0.247% demonstrated superior smoothing but higher erythema (redness) incidence. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A larger controlled study of the same 0.104% and 0.247% retinol creams that finds hydration falling or TEWL rising after 12 weeks, a difference between the two concentrations on those measures, or no higher erythema incidence with 0.247% would show this wrong.
Del Rosso and colleagues reported in The Journal of Clinical and Aesthetic Dermatology that in a lab test mixing equal parts of one optimized water-based tretinoin (a prescription retinoid) gel (0.05%) and a 6.26% benzoyl peroxide (an acne medicine) gel at 32°C, 100 percent of the starting tretinoin remained after seven hours; they called the direct clinical significance unknown. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
A repeat lab assay of the same optimized tretinoin gel (0.05%) mixed in equal parts with the 6.26% benzoyl peroxide gel at 32°C that finds measurable tretinoin loss or rising breakdown products within seven hours would show this wrong.
The FDA's page on wrinkle treatments says a product intended, for example, to remove wrinkles or increase the skin's production of collagen is a drug or a medical device. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
An FDA page or rule stating that a product intended to increase the skin's production of collagen can be sold as a cosmetic without meeting drug or device requirements.
The FDA's page on the Modernization of Cosmetics Regulation Act of 2022 says the responsible person (the company accountable for a cosmetic) must keep records that adequately support the safety of its cosmetic products, and that neither the law nor FDA rules require specific tests to demonstrate the safety of individual products or ingredients. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
An FDA page or rule under the Modernization of Cosmetics Regulation Act of 2022 that requires specific safety tests for each cosmetic product, or that drops the duty to keep records supporting product safety.
Zaghloul and colleagues, writing in The British Journal of Dermatology in 2005, checked actual use after 3 months in 403 acne patients on isotretinoin (an oral acne drug) or other treatments, and found use averaged 64.7% of the expected amount. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The paper reports a mean objective adherence other than 64.7%, or a completed sample other than 403 patients.
Yang and colleagues' 2026 survey in Patient Preference and Adherence of patients aged 16 to 29 in China diagnosed with AV (acne vulgaris, the common form of acne) found 50.00% of retinoid users stopped within two weeks. Across topical medicines, the main reasons patients gave for stopping were slow onset or poor efficacy (44.00%) and concerns about or inability to tolerate side effects (40.00%). source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The survey reports a share other than 50.00% of retinoid users stopping within two weeks, or gives main reasons for stopping other than slow onset or poor efficacy (44.00%) and side effects (40.00%).
Shiv and colleagues reported in the Journal of Marketing Research in 2005 that, across three experiments, people who paid a discounted price for a product such as an energy drink may get less actual benefit from it, for example solving fewer puzzles, than people who paid the regular price for the same product. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The paper's three experiments show no difference in benefit, such as puzzles solved, between people who paid the discounted price and people who paid the regular price for the same product.
Circana's 2026 report on U.S. beauty retail says prestige skincare (higher-priced brands) grew 3% in dollar sales in 2025, while mass-market skincare (lower-priced brands) grew 6% in both dollars and units. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
Circana's report gives a 2025 prestige skincare dollar growth other than 3%, or a mass-retail skincare increase other than 6% in both dollars and units.
A 2020 meta-analysis (a pooled analysis of earlier studies) in the Journal of the European Academy of Dermatology and Venereology, by Chen and colleagues, put self-declared sensitive skin, to some degree, at 71% of the adult population worldwide, and very or moderately sensitive skin at 40%, while noting that questionnaire-based studies might overestimate it. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The meta-analysis's pooled self-declared prevalence is not 71% of adults, its figure for very or moderately sensitive skin is not 40%, or it does not say questionnaire-based studies might overestimate prevalence.
The U.S. Food and Drug Administration's prescription-to-nonprescription switch list shows Differin Gel (adapalene, a retinoid acne gel) 0.1% approved for over-the-counter acne use on 7/8/2016. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The FDA switch list gives a different approval date for Differin Gel 0.1%, or does not list it as an acne switch.
The U.S. prescribing information for isotretinoin capsules (an oral acne pill), revised 2026-08, says the drug must not be used in pregnancy because it can cause fetal harm, with a higher risk of major birth defects, miscarriage and premature birth. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The current U.S. isotretinoin capsule label does not list pregnancy as a contraindication, or does not name these risks.
The EU's Scientific Committee on Consumer Safety in 2022 called vitamin A in cosmetics safe at 0.05% Retinol Equivalent (a vitamin A unit) in body lotion and 0.3% in other leave-on and rinse-off products. It reached no conclusion on cosmetics' contribution to overall exposure, but said it may be of concern for the 5% most exposed through food and supplements. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The SCCS opinion says vitamin A in cosmetics is unsafe at 0.05% Retinol Equivalent in body lotion or 0.3% in other products, or raises no concern for the 5% of people most exposed through food and food supplements.
The UK electronic medicines compendium (emc), a drug information site, lists Treclin gel (clindamycin, an antibiotic, with tretinoin, a retinoid) as a prescription only medicine, with its page updated 2026-03-05. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The emc page lists Treclin as anything other than a prescription only medicine.
The U.S. label for azelaic acid gel 15%, updated 2025-12-24, says redness (erythema) eased somewhat in trial patients who had rosacea's papules and pustules (red bumps and pus-filled spots), but its effect on rosacea redness without them has not been evaluated. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The label states that azelaic acid gel 15% was tested and shown to work for rosacea redness without papules and pustules.
The UK electronic medicines compendium (emc) prescribing summary for Differin gel (adapalene, a retinoid acne gel), updated 2026-10-21, says it can be used while breastfeeding but should not be put on the chest then, so the baby does not touch treated skin. source
Checked by Product.ai on 2026-11-04
What would prove this wrong
The Differin prescribing summary on emc says the gel should not be used during breastfeeding, or allows application to the chest while breastfeeding.