One hundred and twenty-five people applied either 0.05% retinaldehyde, 0.05% retinoic acid, or an inactive cream, and were followed for 44 weeks. Rather than relying on visual grading, the researchers took silicone moulds of the crow's feet area and measured the surface with an optical instrument. At week 18, both active creams significantly reduced wrinkle and roughness measurements; the inactive cream did nothing at any point. Two findings deserve more attention than they usually get. Retinoic acid caused more local irritation and this affected whether people kept using it, while retinaldehyde was well tolerated throughout. And by week 44 the improvement was less pronounced in both active groups than it had been at week 18 — a result that complicates the assumption that retinoid benefits simply accumulate over time. The published abstract reports statistical significance but no effect sizes, so any specific percentage attributed to this study did not come from it.
Creidi P, Vienne MP, Ochonisky S, et al. Profilometric evaluation of photodamage after topical retinaldehyde and retinoic acid treatment. J Am Acad Dermatol. 1998;39(6):960-965.