Home › Articles › Alzheimer's or normal aging?
Is it Alzheimer's or normal aging? How to tell the difference
Everyone over fifty forgets names, loses keys and walks into a room for no reason. Alzheimer's starts somewhere inside that ordinary noise, which is why it is missed for years. The difference is not in whether memory slips, but in what kind, how often, and whether the person can work around it.
What normal aging does to memory
The brain changes with age the way the rest of the body does. Processing gets slower. It takes longer to learn something new and longer to pull a name out of storage. Holding two tasks in mind at once gets harder. A word sits on the tip of the tongue and arrives ten minutes later, in the car. None of this is disease.
Three features mark this ordinary kind of change. The information is still there: given a cue, a list of options, or a bit more time, the person gets it. The person notices the lapse, often before anyone else does, and is usually irritated by it. And daily life carries on. The bills get paid, the appointments get kept, the route home is still the route home. Slower is not the same as lost.
Side by side: normal versus worth checking
These pairs describe the same situation twice, once as age-related change and once as the kind of change a memory clinic takes seriously. They are drawn from the clinical guidance summarised on our warning signs page.
Usually normal aging
- Forgetting a name, then remembering it later.
- Misplacing the glasses, then retracing steps and finding them.
- Occasionally forgetting which day it is, and working it out.
- Needing help with a new phone or an unfamiliar app.
- Making an occasional bad call about money.
- Finding it harder to follow a fast, crowded conversation.
- Being set in your ways, or tired of large gatherings.
- Losing the thread of a story now and then.
Worth a medical opinion
- Forgetting recently learned information and not recovering it, or asking the same question repeatedly.
- Putting things in odd places, such as keys in the fridge, and being unable to retrace steps.
- Losing track of the season or the year, or not knowing how you got somewhere.
- Struggling with a task done for decades: a familiar recipe, a well-known game, the car's own controls.
- A run of poor financial decisions, or being taken in by obvious scams.
- Stopping mid-sentence with no idea how to continue, or calling things by the wrong name.
- Withdrawing from work and hobbies, or a marked change in mood, suspicion or personality.
- Family noticing problems the person insists are not there.
The pattern matters more than any one lapse
No single item on the right-hand list is diagnostic. Everyone has had a day like that. What clinicians weigh is four things taken together.
- Frequency. Once a month is noise. Several times a week is a pattern.
- Direction. Normal aging is flat or very slowly changing. A decline that is clearly worse than six months ago, and worse again than a year ago, is the signal.
- Who notices. In ordinary aging the person complains and the family shrugs. In early Alzheimer's that often reverses: the family is worried and the person explains each lapse away. Neither pattern is proof, but the reversal is a reason to look.
- Function. The question a doctor will ask is not "do you forget things?" but "has anything changed in what you can manage on your own?" Managing medicines, money, cooking, driving and appointments are the usual tests. Trouble with these, when they used to be easy, is the single most useful sign.
There is also a stage between normal aging and dementia, called mild cognitive impairment, where testing shows a measurable change but daily life is still managed. It is common, it is not always Alzheimer's, and it does not always progress. We cover what one large cohort shows about it in a separate article.
When the person is the one worrying
Researchers have a term for the experience of feeling your memory is worse while tests still come back normal: subjective cognitive decline. It has been studied a great deal, because it is one of the earliest points at which a problem could be caught.
A meta-analysis in Neuropsychology Review pooled 46 longitudinal studies with more than 74,000 participants. People reporting subjective decline had a hazard ratio of 1.90 for later dementia, with a 95% confidence interval of 1.52 to 2.36, and 1.73 for later mild cognitive impairment. Including worry in the definition, and being recruited from a clinic rather than the community, raised the odds of later MCI.
Two things follow, and the second is as important as the first. A persistent sense that your memory has changed is a real risk marker and deserves to be taken seriously, not waved away. But a hazard ratio under two means the risk is raised, not decided: the outcome for most people with that worry is not fixed by the worry. Our reading of this, which is interpretation rather than a finding of the paper, is that subjective decline earns a proper assessment, not a conclusion.
Two further studies in the library sharpen the picture. A Framingham Heart Study analysis in JAMA Psychiatry followed 3,585 cognitively normal adults aged 60 and over for up to 12 years to test the newer "SCD-plus" criteria, a stricter definition proposed by an international working group, in a community sample rather than a clinic. And a meta-analysis in Neurology asked the question clinics now ask: does subjective decline carry more risk when a biomarker also shows Alzheimer's changes in the brain? Both are listed at the end of this article. The direction of travel in this research is away from the complaint on its own and towards the complaint plus objective evidence.
What is happening in the brain
Normal aging is not the absence of brain change. In the Baltimore Longitudinal Study of Aging, 138 non-demented people aged 64 to 86 were scanned every year for up to ten years, 1,017 scans in all. Every brain region measured shrank with normal aging. The 18 participants who developed mild cognitive impairment during the study showed a distinct pattern of faster loss in specific regions, which the authors describe as a unique structural vulnerability. The difference between aging and disease was not whether the brain changed, but where and how fast.
A second study, in Brain, tracked tau protein with PET scans over 18 months in 202 people spanning normal aging, MCI and Alzheimer's dementia. Cortical tau rose measurably only in participants who already had amyloid in the brain. In amyloid-negative participants it did not move. That is the biological version of the side-by-side table above: the same symptom can sit on two different processes, and only one of them is Alzheimer's. What those two proteins do is the subject of another article.
What to do about a worry
See a doctor, and take someone who knows you well. A primary care doctor can do a brief cognitive screen, review every medicine, and order blood tests. Several common conditions produce memory symptoms and are treatable: thyroid problems, vitamin B12 deficiency, depression, sleep apnoea, untreated hearing loss, and the combined effect of sedating medicines. Finding one of these is a good outcome of the visit, not a wasted trip.
Write down specifics beforehand. "Forgot the grandchildren's names twice in March; got lost driving to the pharmacy in April" is worth more to a clinician than "memory is bad". Note what changed, when it started, and what the person can no longer do alone.
Ask what the next step would be. If the screen raises concern, the route runs through more detailed cognitive testing, a brain scan, and increasingly a blood test for Alzheimer's proteins. Our diagnosis page walks through what that process involves and how long it tends to take.
Do not rely on an online memory test. They cannot separate a bad night's sleep from a disease, and a reassuring score can delay a visit that should happen now.
Do not wait for it to become obvious. The treatments that exist, and nearly all clinical trials, are aimed at the earliest stage. The paperwork that protects a person's wishes also has to be signed while they can still sign it. An early answer, even an unwelcome one, buys choices that a late answer does not.
From the research library
The studies this article draws on, each with its own plain-English page. Scores are alzhub's usefulness score, which rates how much weight a study can bear, not whether its result is good news.
- Longitudinal pattern of regional brain volume change differentiates normal aging from MCI
- Subjective Cognitive Decline: Level of Risk for Future Dementia and Mild Cognitive Impairment, a Meta-Analysis of Longitudinal Studies
- Subjective Cognitive Decline Plus and Longitudinal Assessment and Risk for Cognitive Impairment
- Progression of Subjective Cognitive Decline to MCI or Dementia in Relation to Biomarkers for Alzheimer Disease: A Meta-analysis
- A multicentre longitudinal study of flortaucipir (18F) in normal ageing, mild cognitive impairment and Alzheimer's disease dementia