One in three dementia cases is attributable to modifiable risk factors. That number comes from the 2024 Lancet Commission on dementia prevention — and it means the disease is not simply a coin flip encoded in your genes. What you do in your 40s and 50s is writing the first draft of your cognitive future.

This guide covers what the evidence actually supports, where promising interventions sit on the speculation spectrum, and how peptides and supplements fit (or don't) into a serious prevention strategy. No hype. No miracle cures. Just a clear-eyed look at what might move the needle.

⚕️ Medical disclaimer: This article is educational and does not constitute medical advice. Peptides are not FDA-approved for dementia prevention. Consult a qualified physician before beginning any protocol.

The Modifiable Risk Landscape

The 2024 Lancet Commission identified 14 modifiable risk factors that collectively account for ~45% of dementia cases worldwide (up from 40% in the 2020 update, as education and hearing loss data improved):

Risk Factor	Population Attributable Fraction	Window
Less education (early life)	~5%	Childhood–young adult
Hearing loss	~7%	Midlife
LDL cholesterol (high)	~7%	Midlife
Depression	~3%	Midlife + later life
Physical inactivity	~2%	Midlife + later life
Diabetes	~2%	Midlife + later life
Smoking	~2%	Midlife + later life
Hypertension	~2%	Midlife
Obesity	~1%	Midlife
Social isolation	~5%	Later life
Excessive alcohol	~1%	Later life
Traumatic brain injury	~3%	Later life
Air pollution	~3%	Later life
Vision loss (untreated)	~2%	Later life

The practical implication: high-value targets cluster in midlife (your 40s and 50s), not retirement age. Waiting until symptoms appear is waiting too long — amyloid pathology begins accumulating 15–20 years before clinical diagnosis.

💡 Key takeaway: Addressing cardiovascular risk, hearing, sleep, and physical activity in midlife has the highest leverage. Supplements and peptides are adjuncts to this foundation, not substitutes.

Exercise: The Strongest Single Intervention

No other intervention has a stronger evidence base for dementia risk reduction than aerobic exercise. The mechanism is multi-pathway: BDNF (brain-derived neurotrophic factor) upregulation, hippocampal neurogenesis, improved cerebrovascular perfusion, and glymphatic waste clearance enhancement.

What the trials show

FINGER trial (2015, n=1,260): Multi-domain lifestyle intervention (aerobic + resistance + diet + cognitive training) produced a 25–150% improvement across cognitive domains vs. general health advice over 2 years.
Erickson et al. (2011, PNAS): Randomized aerobic exercise (walking, 3×/week, 1 year) increased hippocampal volume by ~2% vs. ~1.4% shrinkage in stretching controls. BDNF rose correspondingly.
Meta-analysis (Livingston et al., Lancet 2024): Physical inactivity carries an odds ratio of ~1.4 for dementia — comparable to smoking.

Minimum effective dose

Type	Target	Evidence Level
Aerobic (moderate intensity)	150 min/week (e.g., 30 min × 5 days)	Strong (multiple RCTs)
Vigorous aerobic	75 min/week equivalent	Strong
Resistance training	2×/week	Moderate (additive benefit)
Zone 2 cardio specifically	3–4 hrs/week (emerging)	Observational; no RCT yet

The practical priority: Consistent moderate aerobic activity beats optimized "biohacking" protocols done sporadically. Walk fast, cycle, swim — frequency matters more than intensity for this endpoint.

Sleep: The Brain's Waste-Clearance System

Sleep isn't rest. It's when the glymphatic system — a network of channels that clears metabolic waste — runs its nightly flush. This system is ~60% more active during sleep than wakefulness, and amyloid-beta is one of its primary substrates.

A single night of sleep deprivation measurably increases amyloid-beta in cerebrospinal fluid (Shokri-Kojori et al., PNAS 2018). Chronic short sleep (<6 hours/night in midlife) is associated with a 30% increased dementia risk in a 25-year follow-up cohort (Sabia et al., Nature Communications 2021).

What actually improves sleep architecture

Consistent bedtime/wake time: Circadian entrainment. The single highest-leverage behavioral change for sleep quality.
Temperature: 65–68°F (18–20°C) bedroom. Core temperature drop initiates sleep onset.
Light hygiene: Blue light suppresses melatonin onset for ~3 hours. Evening dimming matters.
Alcohol: Even modest amounts fragment slow-wave and REM sleep — the stages most relevant to glymphatic clearance.
CBT-I: Cognitive Behavioral Therapy for Insomnia outperforms sleep medications for chronic insomnia in RCTs. First-line treatment per American College of Physicians.

⚠️ Sleep apnea is underdiagnosed and high-stakes. Untreated OSA fragments sleep architecture, reduces glymphatic clearance efficiency, and is independently associated with accelerated amyloid accumulation. If you snore or wake unrefreshed, test for OSA before optimizing supplements.

Nutrition: The MIND Diet Evidence

The MIND diet (Mediterranean-DASH Intervention for Neurodegenerative Delay) was designed specifically to protect brain health, combining elements of the Mediterranean and DASH diets. It is the best-studied dietary pattern for cognitive outcomes.

MIND diet components

Food Group	Frequency	Primary Benefit
Leafy greens (kale, spinach, collards)	6+ servings/week	Vitamin K, lutein, folate
Other vegetables	Daily	Antioxidants, fiber
Berries (blueberries, strawberries)	2+ servings/week	Flavonoids, anthocyanins
Nuts	5+ servings/week	Vitamin E, healthy fats
Olive oil	Primary cooking oil	Oleocanthal (anti-inflammatory)
Whole grains	3+ servings/day	Metabolic stability
Fish (fatty)	1+ meal/week	DHA/EPA (omega-3s)
Beans/legumes	4+ meals/week	Protein, fiber
Poultry	2+ meals/week	Lean protein
Wine (red)	≤1 glass/day	Resveratrol (modest evidence)

The evidence: Morris et al. (2015, Alzheimer's & Dementia) showed high MIND diet adherence was associated with cognitive aging equivalent to being 7.5 years younger. A 2023 PREDIMED-Plus sub-analysis showed similar results. Note: these are observational. The ongoing MIND trial (RCT) will provide causal evidence; preliminary results are expected by 2026.

What to limit: Red meat, butter/margarine, cheese, pastries, sweets, fried/fast food. These are the MIND diet's explicit "avoid" categories, associated with higher cognitive risk in cohort data.

Cognitive reserve: the concept that matters

Cognitive reserve is the brain's resilience against pathology — built over a lifetime through education, complex occupation, learning, and social engagement. Higher reserve delays symptom onset even when amyloid burden is substantial. Post-mortem studies routinely find individuals who met pathological criteria for Alzheimer's but died without dementia symptoms; they invariably have higher proxy measures of reserve.

What builds reserve:

Learning a new skill or language (stronger effect than crossword puzzles)
Musical instrument practice (bi-directional training of auditory and motor cortex)
Complex, engaging work (not repetitive task work)
ACTIVE10 trial: structured learning produced measurable hippocampal volume preservation over 2 years

Social isolation carries a population attributable fraction of ~5% for dementia — higher than diabetes, obesity, and physical inactivity individually. Mechanistically: social interaction drives cognitive engagement, reduces cortisol chronically, and may directly stimulate BDNF. Loneliness correlates with accelerated biological aging in multiple methylation clock studies.

The underrated intervention: Hearing aids. Untreated hearing loss is the single largest modifiable risk factor (7% PAF). It drives social withdrawal, cognitive load overload, and reduced stimulation. If you have hearing loss, this is your highest-ROI intervention — ahead of any supplement or peptide.

Supplements: What Has Evidence vs. What Sounds Good

The supplement market for cognitive health is enormous and largely unregulated. Here's a clear-eyed assessment of the four most commonly recommended options.

Supplement	Evidence Quality	Effect Size	WellSourced Verdict
*Lion's Mane (Hericium erinaceus)*	🟡 Limited (small RCTs, animal data)	Modest in MCI patients	Worth trying; not a slam dunk
Omega-3s (DHA/EPA)	🟡 Mixed RCT data; strong mechanistic	Modest for prevention; better for cardiovascular risk reduction	Reasonable adjunct; prioritize dietary fish first
Citicoline (CDP-choline)	🟡 Several RCTs, mostly European; small samples	Modest; better in post-stroke and vascular dementia contexts	Low risk, plausible mechanism; acceptable add-on
Phosphatidylserine (PS)	🟡 Multiple RCTs (soy-derived); FDA qualified health claim (2003)	Modest in MCI and normal aging	Reasonable; set expectations appropriately

Lion's Mane

The most-hyped nootropic of the past five years. Mechanistically interesting: hericenones and erinacines stimulate Nerve Growth Factor (NGF) synthesis, with NGF playing a role in cholinergic neuron survival.

Evidence: Mori et al. (2009) RCT (n=30, MCI patients) showed significant improvements in MMSE scores vs. placebo over 16 weeks — with full reversal after stopping. The 2023 Australian RCT (n=41) found no significant difference on primary cognitive outcomes, though secondary biomarkers improved. The honest summary: it works in some studies, not others; effect sizes are modest; it's well-tolerated. It's not the miracle the wellness internet claims, but it's not nothing either.

Dosing used in trials: 1–3g standardized extract (containing ≥25% β-glucans) daily. Whole mushroom powder doses are typically 3–5g.

Omega-3s (DHA/EPA)

DHA is the dominant fatty acid in neuronal cell membranes. Its concentration in the brain correlates inversely with dementia risk in observational data. The problem is that this hasn't translated cleanly to RCT outcomes.

OMEGA-AD trial: 18-month supplementation in Alzheimer's patients — no benefit on primary endpoint. MIDAS trial: Algae-derived DHA improved learning and memory in healthy older adults with age-related cognitive decline. Overall: The data supports benefit for prevention in people with low baseline DHA (non-fish-eaters), less so for those already eating fatty fish 2–3×/week. Cardiovascular risk reduction via omega-3s is mechanistically linked to brain health (reduced vascular dementia risk).

Dosing: 1–2g combined EPA+DHA daily. Triglyceride form (fish oil or algae oil) has better bioavailability than ethyl ester form.

Citicoline

CDP-choline is a precursor to phosphatidylcholine and acetylcholine. It crosses the blood-brain barrier and supports membrane repair and cholinergic signaling — both relevant in Alzheimer's pathology.

Citicoline has the most consistent evidence base of any nootropic supplement in vascular and post-stroke cognitive impairment settings. For healthy aging prevention, the data is thinner but mechanistically coherent. Low risk profile. Often stacked with racetams (though racetam evidence is itself limited).

Dosing: 250–500mg daily. Twice-daily dosing used in most trials.

Phosphatidylserine

PS is a phospholipid concentrated in neuronal membranes, involved in cell-to-cell communication and neuronal apoptosis regulation. The FDA issued a qualified health claim in 2003 that "very limited and preliminary scientific research suggests that phosphatidylserine may reduce the risk of dementia in the elderly."

The key word is "qualified." Older trials used bovine-derived PS (no longer available); modern soy-derived PS has modest evidence. Effects are most pronounced in early cognitive decline rather than prevention in healthy adults.

Dosing: 100–300mg daily, often taken with meals.

Peptides: Semax, Selank, Dihexa, BPC-157

This is where we enter genuinely speculative territory — but not without reason to look closely.

🔔 Regulatory status: None of these peptides are FDA-approved for cognitive indications in the United States. Semax and Selank are approved drugs in Russia. These are research chemicals in the US context. The evidence below is real, but the clinical translation gap is large.

Peptide	Primary Mechanism	Evidence Status	Risk Level
Semax	BDNF/ACTH analogue → BDNF upregulation, neuroprotection	🔬 Human trials (Russia, stroke/TBI); preclinical cognitive enhancement	Low — short-acting nasal spray
Selank	Tuftsin analogue → anxiolytic, BDNF modulation, IL-6 reduction	🔬 Human trials (Russia, anxiety/PTSD); cognitive secondary outcomes	Low — nasal spray
Dihexa	HGF activator → Met receptor → synaptic strengthening	🔬 Animal data only; potency claims are extraordinary	Unknown — no human safety data
BPC-157	Gastric peptide → VEGF, GABA, dopamine modulation; anti-inflammatory	🔬 Primarily animal; some indirect human relevance via gut-brain axis	Low-moderate (extensive animal safety)

Semax: The Most Promising in the Cognitive Category

Semax is a heptapeptide analogue of ACTH(4-10) developed in Russia, where it's been used clinically for stroke recovery and cognitive enhancement since the 1980s. Its primary mechanism is upregulation of BDNF and its receptor TrkB in the prefrontal cortex and hippocampus.

What the evidence shows:

Multiple Russian clinical trials in stroke rehabilitation demonstrate improved cognitive recovery
Preclinical studies show protection against beta-amyloid toxicity in rodent models
BDNF upregulation is one of the most reproducible effects in peptide literature — and BDNF is a credible neuroprotective target
No large Western RCTs exist for Alzheimer's prevention specifically

Honest assessment: Semax has the most credible mechanism and most clinical data in this category. If you're going to experiment with a cognitive peptide, this has the best risk-benefit ratio. The nasal spray formulation avoids injection. But recognize you're operating in uncharted territory for prevention — the clinical evidence is for acute recovery, not long-term prophylaxis.

Dosing used in trials: 300–600mcg intranasally, 1–2×/day. Typical cycles: 2–4 weeks on, 1–2 weeks off.

Selank: Anxiolytic-First, Cognitive Second

Selank is a synthetic analogue of the human peptide tuftsin. Its primary use in Russian clinical practice is anxiety and stress — and this is relevant to cognition because chronic stress is one of the most potent accelerators of hippocampal atrophy and cognitive aging.

Selank reduces IL-6 and other pro-inflammatory cytokines, modulates serotonin and dopamine tone, and appears to stabilize BDNF levels under stress conditions. If you have significant stress-driven cognitive symptoms (brain fog, anxiety, poor focus), the mechanism is plausible. For dementia prevention specifically, the evidence is too indirect to make a strong claim.

Dihexa: High Potency Claims, Zero Human Data

Dihexa (PNB-0408) received significant attention after a 2013 paper from Washington State University showed it was approximately 10 million times more potent than BDNF at stimulating synaptic connections in rodents. It activates the HGF/Met receptor pathway, driving synaptogenesis.

The critical caveat: No human safety or efficacy data exists. HGF/Met activation is implicated in cancer progression — it's the same pathway targeted by cancer drugs trying to inhibit it. Until there are Phase I human safety trials, using Dihexa is taking an unknown risk. The potency claims make it simultaneously intriguing and concerning. WellSourced's position: don't use it yet.

BPC-157: The Gut-Brain Connection

BPC-157 is the peptide most of this community already knows from recovery and injury applications. Its cognitive relevance comes from the gut-brain axis — gut dysbiosis and neuroinflammation are increasingly recognized in Alzheimer's pathology — plus its direct effects on dopamine and GABA systems in the CNS.

The honest picture: BPC-157 is one of the most extensively studied peptides in animal models (over 200 studies) but has essentially no completed human RCTs. Its neuroprotective effects in rodent models are real. Whether they translate meaningfully to human dementia prevention is unknown. It is reasonable to use for recovery and joint health — its potential cognitive benefit is a secondary bonus with no direct human evidence. See our BPC-157 Protocol Guide for full dosing and stacking guidance.

Putting It Together: A Practical Protocol

Priority order matters here. The highest-leverage interventions are behavioral, not supplemental. Here's how to stack them:

Tier 1: Non-Negotiable Foundation

✅ 150+ min/week aerobic exercise (walk, cycle, swim — zone 2 or moderate intensity)
✅ 7–9 hours sleep with consistent schedule; treat OSA if present
✅ Control hypertension — systolic <130 mmHg; strongest late-life modifiable risk
✅ Hearing health — use hearing aids if needed; this is the highest-PAF modifiable factor
✅ MIND diet pattern — especially leafy greens, berries, fish, olive oil

Tier 2: Evidence-Supported Adjuncts

🟡 Omega-3s: 1–2g DHA+EPA/day if dietary fish intake is low
🟡 Citicoline: 250–500mg/day — choline support, membrane health
🟡 Lion's Mane: 1–3g extract/day — NGF support; modest evidence, low risk
🟡 Phosphatidylserine: 100–300mg/day — especially if early cognitive symptoms
🟡 Social engagement: Active community involvement, new skill acquisition

Tier 3: Speculative / High-Curiosity

🔬 Semax (300–600mcg IN): Most credible peptide option; use cyclically, not daily
🔬 Selank (250–500mcg IN): If stress-driven cognitive symptoms are significant
🔬 BPC-157: If already using for recovery — cognitive benefit is plausible secondary effect. See BPC-157 Protocol Guide
❌ Dihexa: Avoid until human safety data exists

For a broader peptide ranking by evidence, see the 2026 Peptide Tier List. If you're 50+ and thinking about the cognitive angle specifically, the Best Peptides for Men 50+ and Best Peptides for Women 50+ guides address age-appropriate stacking in more detail. For supplement longevity stacks beyond peptides, the NAC & Tru Niagen stack guide is worth reading alongside this one.

What to monitor

Blood pressure: Home monitoring, quarterly
Fasting glucose / HbA1c: Metabolic health is vascular health is brain health
Omega-3 index: Fingerprick test (target >8%)
Sleep quality: Oura/WHOOP or clinical polysomnography if OSA suspected
Cognitive baseline: CNS Vital Signs or Cambridge Brain Sciences — annual tracking catches trends early

💡 The honest bottom line: No supplement or peptide comes close to the protection conferred by aerobic exercise, quality sleep, blood pressure control, and genuine social engagement. The biology is clear on this. Supplements and peptides layer on top of that foundation — they do not replace it.

Frequently Asked Questions

Is dementia actually preventable, or is it mostly genetic?

For most people, genetics are a risk modifier, not a sentence. The APOE4 variant roughly doubles or quadruples risk, but only ~25% of the population carries one copy, and even APOE4 carriers benefit substantially from lifestyle modifications. The 2024 Lancet Commission estimates 45% of cases are attributable to modifiable risk factors — meaning the majority of dementia is theoretically preventable if those factors are addressed. Genetic risk is real; it is not destiny.

When should I start a prevention protocol?

Now. Amyloid pathology begins accumulating 15–20 years before clinical symptoms. If you're in your 40s or 50s, the interventions with the strongest evidence (exercise, sleep, blood pressure control) are most impactful during this window. There is no age at which starting is too late — the brain retains plasticity into advanced age — but starting earlier compounds the benefit.

Does Lion's Mane actually work?

It has a credible mechanism (NGF stimulation via hericenones/erinacines), has shown benefit in several small RCTs in mild cognitive impairment, and is well-tolerated. However, the effect sizes are modest, results across trials are inconsistent, and no large Phase III trial exists. It's a reasonable, low-risk addition to a cognitive stack — but the wellness internet dramatically overstates the evidence. It should complement, not anchor, your strategy.

What's the difference between Semax and Selank for cognitive health?

Semax is primarily a BDNF stimulator and neuroprotective agent — its evidence is stronger for acute recovery (stroke, TBI) and general cognitive enhancement. Selank is primarily anxiolytic, with cognitive benefits flowing from reduced chronic stress and neuroinflammation. If cognitive enhancement is your primary goal, Semax has better direct evidence. If anxiety and stress-driven brain fog are your main complaints, Selank is more targeted. They can be cycled independently or used together.

Is the MIND diet actually proven to prevent Alzheimer's?

The strongest published evidence is observational (large cohort studies showing lower dementia incidence with higher MIND diet adherence). Causal proof awaits the MIND trial RCT, which is currently ongoing. That said, the MIND diet's components overlap extensively with cardiovascular health interventions that have mechanistic evidence for brain protection — it's not speculative, it's awaiting gold-standard confirmation. Given low risk and multiple overlapping benefits, following it is sensible while we wait for the RCT data.

Should I get genetic testing (APOE4) to know my risk?

This is a personal decision with psychological implications. Knowing you carry APOE4 doesn't change the recommended interventions (they're the same regardless) but it can motivate stronger adherence — or cause significant anxiety. If you're the type who acts on data, it may be useful. If you're prone to health anxiety, knowing may do more harm than good. Discuss with a genetic counselor or physician, not a supplement company. Services like 23andMe report APOE4 status; you can also request ApoE genotyping through clinical labs.

How does BPC-157 connect to brain health?

BPC-157's CNS effects operate through several pathways: it modulates dopamine and serotonin systems, has demonstrated neuroprotective effects in rodent models of traumatic brain injury and stroke, and its anti-inflammatory properties may benefit the gut-brain axis (gut dysbiosis is increasingly implicated in Alzheimer's neuroinflammation). However, all of this evidence is animal-model based — there are no human RCTs on BPC-157 for cognitive outcomes. It's plausible, not proven. If you're using it for joint recovery, the potential brain benefit is a reasonable secondary consideration.

What's the most underrated dementia prevention intervention?

Hearing aids for people with hearing loss. The evidence is among the strongest of any modifiable factor (7% population attributable fraction — the highest of all 14 Lancet Commission risk factors), the mechanism is clear (reduced cognitive load, maintained social engagement, auditory stimulation), and it's completely non-pharmacological. Yet it's vastly underutilized — only about 20% of people who could benefit from hearing aids actually use them. If you have hearing loss, this is step one — before any supplement, before any peptide.

WellSourced covers evidence-based wellness protocols for the research-curious. For more on neuroprotective peptides, see the 2026 Peptide Tier List and Best Peptides for Men 50+.

Dementia Prevention: The Wellness Protocols With Real Evidence (and the Ones That Are Hype)

The Modifiable Risk Landscape

Exercise: The Strongest Single Intervention