Your 40s, evidence first.

The 2025 meta-analysis pools eight RCTs, 16 weeks to 2 years: fine wrinkles improved by a mean 0.41 versus vehicle, coarse wrinkles 0.25, patient self-assessment favoured tretinoin, and irritation was about three times more likely. The age match is unusually good — Weiss 1988 enrolled 35–70, Kang 2005 enrolled 40–76, and two 24-week multicentre trials of 0.02% tretinoin cream enrolled adults 43–70. That 0.02% data makes the "low dose, every night, for years" approach defensible; effects are real and modest — texture and fine lines respond, deep static folds do not.

Caveat: the pivotal trials were manufacturer-funded. Expect 6–8 weeks of irritation; judge at 6 months.

Hughes 2013 randomized 903 Australian adults under 55 to daily SPF 15+ versus discretionary use for 4.5 years and measured skin aging by blinded microtopography: daily users showed no detectable increase in aging, 24% less than the discretionary group (relative odds 0.76). Publicly funded, and the only RCT of any topical with a photoaging endpoint. In the 40s it also guards the treatments below — every IPL result recurs with sun, and every retinoid raises photosensitivity.

Humbert 2003 ran a 6-month double-blind, split-side RCT of 5% vitamin C cream versus excipient in women with photoaged skin and found clinical, profilometric and ultrastructural improvement. Consistent small trials, modest effect, and a stability problem — L-ascorbic acid oxidizes, so opaque airless packaging and regular replacement matter. An adjunct under sunscreen, not a pillar.

Bissett 2005 is rare in matching this decade precisely: 50 white women aged 40–60, 5% niacinamide in a moisturizer versus the moisturizer alone, split-face, 12 weeks — significant improvements in fine lines, hyperpigmented spots, red blotchiness and sallowness. Well-run but one manufacturer's programme (Procter & Gamble). Texture and pigment benefits, not structure — and a useful ceramide booster for the dryness that arrives with perimenopause (see our ceramides guide).

Falling and fluctuating estrogen reduces barrier lipids while sebaceous activity persists — the reason skin can be dry and breaking out at once. Ceramide- and hyaluronic-acid-based moisturizers have consistent trial data for hydration and transepidermal water loss (graded in full in our ceramides guide), and they double as the tolerance tool that keeps a nightly retinoid viable. Expect comfort and glow, not fewer wrinkles.

The biology is plausible — fibroblasts carry estrogen receptors — and the flagship study is seductive: Schmidt 1996 applied 0.01% estradiol or 0.3% estriol cream to 59 perimenopausal women for six months and reported markedly improved elasticity and firmness, reduced wrinkle depth and pore size, and more type III collagen on biopsy. The catch: the two actives were compared with each other, not with placebo, so the headline numbers are uncontrolled. A topical estrogen-receptor agonist has a double-blind pilot. No licensed facial estrogen cream exists in Europe; it is compounded, off-label, and its systemic absorption from the face is poorly characterized. Interesting; not yet a recommendation.

The evidence retrieved for peptides is in vitro — copper tripeptide increased fibroblast growth-factor secretion in culture — and topical growth factors rest on open-label industry work. Neither has an RCT with clinical photoaging endpoints. Pleasant, plausible, and priced as if proven; the money buys more skin in tretinoin and sunscreen.

The 2022 Menopause Society position statement is unambiguous: hormone therapy remains the most effective treatment for vasomotor symptoms and genitourinary syndrome, prevents bone loss and fracture, and for women under 60 or within 10 years of menopause onset without contraindications the benefit–risk ratio is favourable; beyond that window it worsens. Skin is not a listed indication — but everyone reading this decade's guide is inside the favourable window by definition, and it closes with time rather than with age alone.

Practical trigger: cycle irregularity plus any of hot flashes, night sweats, new insomnia, new acne or hair shedding, low mood or vaginal symptoms — or simply to plan. A society-certified menopause specialist is the safeguard against both under-treatment and "optimization" clinics.

Not a skin treatment, but on any honest list for this decade: low-dose vaginal estrogen is first-line for genitourinary syndrome of menopause in the same position statement, with minimal systemic absorption and few contraindications. Dryness, discomfort and recurrent urinary symptoms are common from the late 40s and widely under-treated.

For women who can't or won't take hormones, the neurokinin-3 antagonist fezolinetant has a real trial base: SKYLIGHT 1 (527 women aged 40–65 at 97 sites including several European countries; Astellas-funded) showed significant reductions in vasomotor frequency and severity over 12 weeks with a 40-week extension. Skin relevance is indirect but real — hot-flash control is the evidence-based route to unbroken sleep in perimenopause, and sleep loss shows up on intrinsic-aging scales.

A 2023 network meta-analysis of 19 RCTs (919 women) found moderate-intensity resistance training improved lumbar-spine and femoral-neck density in postmenopausal women, and aerobic-only training consistently underperforms it for bone. The rationale for starting in the 40s is the SWAN timeline: lean mass and bone begin falling one to two years before the final period. Protein at ~1.2 g/kg/day with training is the guideline floor — the 1.6 g/kg figure is athletic-population extrapolation, and one RCT found higher protein added nothing to lean-mass gains over the RDA.

Two guideline-backed tests fit this decade. European consensus recommends measuring Lp(a) at least once in every adult (risk rule-in ≥125 nmol/L), and cardiovascular risk climbs after menopause — the 40s are the natural moment. For bone, USPSTF screens all women at 65 and younger postmenopausal women by fracture-risk score; a baseline DEXA at the final period, or in the late 40s with risk factors (low BMI, early menopause, steroids, parental hip fracture, smoking), is defensible — routine DEXA at 40 is not. Add lipids with ApoB, HbA1c and thyroid, since both diabetes and thyroid disease mimic menopausal symptoms.

Dry skin and new breakouts arriving together is classic perimenopause — estrogen falls while androgen effect is relatively unopposed. The SAFA trial (pragmatic, double-blind, NIHR-funded, 410 women, spironolactone 50→100 mg/day for 24 weeks) showed a clear quality-of-life and investigator-rated benefit; the mean age was 29, so the 40s are an extrapolation, though the drug is routinely used off-label for exactly this pattern. Niacinamide plus a ceramide moisturizer handles the dryness half.

A quarter of women have detectable pattern hair loss by 49. Bloodwork first (iron, thyroid, a shedding history), then treatment with real trials: low-dose oral minoxidil 1 mg matched topical 5% for hair density at 24 weeks in an RCT (more hypertrichosis, dose-dependent), and a meta-analysis found spironolactone combination therapy improved 66% of women versus 43% on monotherapy. Add microneedling if you want the strongest adjunct data (see our microneedling guide).

Creatine plus resistance training beats training alone for strength in older women, especially over 24-plus weeks. Tempering the bone hopes: the 2-year RCT in 237 postmenopausal women found no effect on bone density, though it preserved femoral-neck geometry — a strength surrogate. Perimenopause-specific data are an 8-week pilot with cognitive endpoints. Cheap, safe, useful for strength; not a bone or skin drug.

Pooled small studies report that HT increases skin elasticity, collagen and thickness (2023 meta-analysis) — but the trials are small, heterogeneous and mostly old, and the largest RCT measuring the face, Phillips 2008 (485 postmenopausal women, 48 weeks of low-dose estrogen/progestin), showed no meaningful wrinkle benefit over placebo. Reviews conclude the role is supportive, and no guideline supports HT for skin alone. Take it for symptoms and bone if you qualify; count thicker, better-hydrated skin as a plausible bonus.

The only objective data: 20 patients imaged before and after GLP-1 therapy (mean loss 11 kg) lost a median 9% of midface volume — superficial fat more than deep — with loss proportional to weight lost. Tiny, retrospective, no non-drug comparator: this is weight loss, not a drug effect. What to do: keep loss to ≤0.5–1 kg a week, resistance-train and hit protein to protect lean mass, wait until weight is stable before any filler or biostimulator, then treat conservatively. Claims that hyperdilute CaHA "prevents" facial volume loss during GLP-1 use rest on early open-label data.