When Memory Changes…Why Early Assessment Matters More Than Ever

September is World Alzheimer’s Month.

September is World Alzheimer’s Month, and this year the focus is on early diagnosis.

I have been thinking about what that actually means.

Sometimes I look at the bookshelf of someone I love but am losing due to dementia. It is full of books she has read over a lifetime. Books that interested her, changed her and probably became part of who she was. They are all still there. But do they still shape her in any sense?

I find that thought almost unbearable.

If our memories disappear, and with them much of our personality, what is left? We say that the person is still there, and of course their human value remains. But what do we actually mean by the person? Is it the body? Is there something separate from memory and personality? Or do we insist that the person is still there partly because the alternative is too painful?

Sometimes severe dementia can make someone we love feel like an emptying thermos. The familiar shape remains, we sense the volume left by it weight, but we no longer really know how much of the person we knew is still inside. It is a cruel thought, perhaps even an unfair one. I suspect many relatives have had it and then felt ashamed.

Most people who worry about their memory are not really frightened by forgetting a name or misplacing their keys. They are frightened by where it might lead.

Until quite recently, medicine did not have a very good answer to that fear. We could investigate, sometimes explain, and offer support and symptomatic treatment. But if the cause was Alzheimer’s disease, we had very little that could alter what would happen next.

This is beginning to change. Slowly, and with more complications than the headlines usually suggest, but it is changing.

Not every memory problem is Alzheimer’s disease

Memory is affected by almost everything.

It works less well when we are tired, distracted, anxious or depressed. Sleep problems, stress, menopause, pain, alcohol, thyroid disease, vitamin deficiencies and several medications can affect concentration and recall. Hearing loss can look like memory loss when the information was never heard clearly enough to be stored in the first place.

This is why an assessment cannot start with a biomarker. It has to start with the person.

What has changed? When did it begin? Is it getting worse? Who noticed it first? Does it affect work, finances, medication, driving or ordinary routines? Has sleep changed? Mood? Movement? Language? What does someone close to the person see?

I am repeating the word person deliberately. A laboratory result can tell us something important about biology. It cannot tell us who is sitting in front of us, what has changed in their life or what that change means to them.

The story matters. The neurological examination matters. Cognitive testing, brain imaging and ordinary blood tests still matter.

The new blood tests

One of the most interesting developments is that we can now detect signs of Alzheimer’s pathology in an ordinary blood sample.

Alzheimer’s disease is associated with changes in amyloid and tau proteins in the brain. Until recently, demonstrating this usually required either a lumbar puncture to analyse cerebrospinal fluid or an amyloid PET scan. Both are useful, but neither is especially simple or widely available.

We can now measure proteins in blood that are closely related to this process. The most promising at present is phosphorylated tau 217, or p-tau217. In May 2026, a p-tau217 test received CE marking in Europe. The result can indicate whether amyloid pathology is likely or unlikely to be present.

This could make assessment considerably easier. It may help us decide which patients need further testing and could reduce the number of lumbar punctures and PET scans.

But a blood test for Alzheimer’s pathology is not the same thing as a diagnosis of dementia.

It does not tell us whether memory problems are affecting daily life. It does not explain every cause of cognitive difficulty, and it is not intended as a screening test for healthy people who are curious about their future. A positive result still has to be understood together with the history, cognitive assessment and other findings. Some results will remain uncertain and require further investigation.

I can imagine that these tests will soon become much easier to request. That is exciting, but also slightly worrying. Once information becomes easy to obtain, we sometimes forget to ask whether it will actually help the person receiving it.

Before taking a biomarker test, we should know what question we are trying to answer. We should also have some idea of what we will do with the result.

Treatments that were not available before

There is now another reason why an early diagnosis may matter.

Lecanemab, sold as Leqembi, and donanemab, sold as Kisunla, are antibodies that remove amyloid from the brain. They do not cure Alzheimer’s disease. They do not bring back memories that have already been lost, and they do not stop the disease. In studies, they slowed cognitive and functional decline to a modest degree in selected patients with early disease.

The treatments are approved for people with mild cognitive impairment or mild dementia caused by Alzheimer’s disease, and amyloid pathology must be confirmed. In Europe, treatment is restricted to people with no copy or one copy of the ApoE4 gene. People with two copies have a higher risk of ARIA, a treatment complication that can cause swelling or bleeding in the brain.

So this is not simply a prescription. Patients need genetic testing, MRI scans before and during treatment, repeated intravenous infusions and careful follow-up. Other illnesses, anticoagulant medication and findings on previous brain scans may make treatment unsuitable.

Leqembi received EU marketing authorisation in April 2025 and Kisunla in September 2025. This is a real change. For the first time in Europe, we have treatments that affect part of the underlying disease process rather than only its symptoms.

Still, approval does not necessarily mean access. In July 2026, Spain decided not to include either treatment in public funding. For the moment, this limits access largely to private care.

I find this difficult. The benefits are modest and the treatments are expensive and demanding. The risks are real. Healthcare resources are not unlimited. At the same time, it is difficult to explain to a patient that a treatment has been approved, may slow the loss of independence, but is not available to them through the public system.

There is no simple answer to that. Pretending that the drugs are a miracle would be dishonest. Pretending that a modest slowing has no value would also be dishonest. The value of time depends very much on whose time it is.

So why investigate early?

Some people avoid assessment because they are afraid of the diagnosis. Others believe there is little point because nothing can be done. I understand both reactions.

But memory problems do not always mean Alzheimer’s disease. An assessment may reveal something treatable. It may show that the problem is not progressive or not primarily neurological. Sometimes it provides reassurance, although reassurance is only useful when it is based on having properly understood the problem.

If the cause is Alzheimer’s disease, knowing earlier may now allow treatment for some patients. It also gives a person more time to make decisions, organise practical matters and say clearly what they want for the future. Families may be able to replace a growing but unspoken worry with some kind of plan.

This year’s World Alzheimer’s Month uses the message The Earlier You Know, The More You Can Do: A Dementia Diagnosis Matters. I think that is true. But only if what follows the diagnosis is thoughtful, honest and available.

A diagnosis without support may simply give fear a name and a biomarker result without context may create more uncertainty than it removes.

Still, we are in a different place than we were only a few years ago. A blood test may soon make the first stages of investigation easier. For some carefully selected patients, treatment may slow the disease and buy time.

And perhaps time is what brings me back to the bookshelf.

The books are still there, exactly where they have always been. But the real library was never the shelves. It was what the books created inside the person who read them: thoughts, associations, opinions, memories, perhaps small changes in personality that accumulated over a lifetime.

That may be part of what we are trying to preserve.

We cannot put back every book once it has disappeared. But if earlier diagnosis and treatment can keep the library open a little longer, that is not nothing.

Dr Jessika Nystedt