A guide to memory supplements: what is in them and what the evidence says

Almost nobody in this market is treating a diagnosis. The buyer is typically between forty and seventy, notices that names arrive slower than they used to, and wants the everyday version of memory back rather than a medical intervention.

That is worth naming because it sets the bar. The outcome measures in the trials behind these ingredients are things like memory free recall, choice reaction time, sustained attention during a demanding task, and composite fatigue indices. They are not measures of dementia risk and they are not measures of intelligence. A supplement that moves one of them by a standardised mean difference of 0.3 has done something real and small.

The broadest published look at whether any of this prevents decline is the Annals of Internal Medicine review, which reviewed the over-the-counter supplement evidence and concluded it was insufficient to recommend any of them. the review of cognitive ingredients in healthy adults asked the narrower question about optimising cognitive performance in healthy adults, and Cicero's review of botanicals and cognitive decline surveyed the botanical half of the field specifically. None of the three is enthusiastic and none of them says nothing works.

Because the ingredient with the best memory evidence, bacopa, was measured at twelve weeks in every trial that found anything, and the reviews that pooled those trials inherited the same window.

That is not a marketing convention imported into the research. It runs the other way: researchers chose twelve weeks because the proposed mechanisms are gradual, and because a shorter trial of a slow intervention finds nothing and teaches nobody anything.

The consequence for a buyer is that almost every review in this category is written too early to mean anything. Review software asks for a rating between one and four weeks after delivery. A one-star review at day 12 and a five-star review at day 12 are both reviews of a fortnight's hope.

The practical response is a long enough guarantee and the discipline to use it. A thirty-day window on a product whose lead ingredient needs ninety days is not really a guarantee at all, and it is worth checking before paying rather than afterwards.

the National Institute on Aging's guidance on memory and ageing is the plainest guide to which memory changes are part of ordinary ageing and which are not, and it is worth ten minutes before any purchase in this category.

Direct-response supplement marketing has a grammar and it is learnable in about five minutes.

A struck-through price that never moves. A permanently discounted price is the normal price. Judge a product against what else is on the shelf rather than against the vendor's own crossed-out figure.

A proprietary blend. A blend name and one total weight makes any dose comparison impossible, which is usually the point. A panel with individual amounts is the single most useful thing a label can do.

Twelve ingredients on a two-capsule serving. There is a fixed weight of powder in a capsule. Twelve names across it means very little of each, and the list exists to be read rather than to be effective.

An unnamed formulator. “Doctor formulated” with no clinician, credential or institution behind it borrows medical authority without accepting any accountability. A named formulator is a checkable claim; a stethoscope is a design element. the FDA's questions and answers on dietary supplements sets out what a manufacturer is actually obliged to substantiate.