A needed revision for the definition of subjective cognitive decline

Subjective cognitive decline (SCD) is a growing area of research in the context of Alzheimer's disease. Indeed, a PubMed search for “subjective cognitive decline” on March 16, 2026 yielded over 6100 citations. A search for titles with “subjective cognitive decline” or subjective cognitive complaints” on the Journal of Alzheimer's Disease webpage resulted in five articles in just these first three months of 2026.^1–5 All of those articles in the journal, including all articles cited in one that was a partial review paper, used definition of SCD developed within the proposed research framework of the SCD-initiative, an international working group of clinicians and researchers.^6,7 As such, my point in the commentary is to call attention to that now nearly universally applied definition rather to any particular study.

The focus of the SCD-initiative was on SCD research as a risk factor for mild cognitive impairment (MCI) and Alzheimer's disease, although they suggested that their SCD criteria could also be useful in healthcare settings. A goal of the SCD-initiative was to provide standardized terminology, but their definition of SCD, which is now widely used, creates two logical/linguistic problems that cause some confusion for researchers, readers, and reviewers of scientific studies. Here I identify those problems and propose a simple modification of terminology to resolve this dilemma.

The SCD-initiative defined SCD on the basis of two features. ⁶ “First, a self-experienced persistent decline in cognitive capacity, compared with a previously normal cognitive status, which is unrelated to an acute event.” Second, “normal performance on standardised cognitive tests used to classify MCI, adjusted for age, sex, and education.” Regarding the second feature, the SCD-initiative also stated the following: “Conditions that are defined by objective cognitive impairment, such as MCI or dementia, are distinct from SCD. Of note, SCD is not a diagnostic category of the International Statistical Classification of Diseases-10, the International Classification of Diseases-11, and the Diagnostic and Statistical Manual of Mental Disorders-5.”

Let's take amnestic MCI as an example to illustrate the source of confusion. By definition, a person diagnosed with amnestic MCI has had objective cognitive—specifically memory—decline. We all know that some people with MCI (and even some with mild dementia) are aware of their cognitive decline, whereas others are not. How are we to describe the experience of those in the former category? In normal English language usage, we would say that they are experiencing SCD. In the research or healthcare setting, this can be useful information, telling us whether or not a person may have some insight into their condition. However, the SCD-initiative criteria leave us without any way of describing this situation. Describing such objectively impaired individuals as experiencing SCD directly contradicts the SCD-initiative definition because it requires the absence of objective cognitive impairment. We could also say that such a person has some awareness or insight into their cognitive status, but that does not solve the problem because not having SCD while being aware of your cognitive decline is clearly illogical and contradictory.

The second problem that arises is that despite the SCD-initiative stating that SCD is not a diagnostic category, requiring the definition of SCD to be restricted to individuals without cognitive impairment does, in effect, make it at least a quasi-diagnostic category. It makes it a quasi-diagnostic category because their criteria unintentionally preclude standard and common English language usage to describe individuals who already have MCI or dementia and are also aware of their own cognitive decline. In common parlance, it would be perfectly natural, acceptable, and correct to refer to those individuals as experiencing SCD, but that description is not permitted according to the SCD-initiative.

To resolve this minor dilemma, I propose a simple revision of the SCD-initiative terminology in which there are two SCD terms: preclinical SCD (P-SCD) and clinical SCD (C-SCD). I propose P-SCD to refer to individuals who are experiencing subjective decline while they remain objectively cognitively normal based on standardized objective testing. I propose C-SCD to refer to individuals who are experiencing actual decline that is enough to constitute objective cognitive impairment based on standardized testing, but who are also aware of having some objective cognitive decline.

I fully realize that being able to use P-SCD to identify people at risk for MCI or dementia while they are still cognitively unimpaired and to possibly intervene to slow the disease process is the primary purpose of SCD research. That goal is of obvious importance. On the other hand, identifying the presence or absence of C-SCD in already impaired individuals may also be of value. Individuals with C-SCD may have a better prognosis and be more amenable to certain interventions than those who have anosognosia, i.e., those who are unaware of or are in denial of their cognitive decline. For individuals with anosognosia, working with them on acknowledging it may be of value for getting them to adhere to relevant treatment and lifestyle changes. For individuals with P-SCD, reassurance that they are not objectively impaired may be helpful, but exploration of what underlies their subjective concern may also be of value. One interesting possibility may be that the experience of decline while test performance remains within the normal range is related to the feeling of requiring greater mental effort to achieve that test performance. Pupillometry, which provides an objective index of mental effort (amount of pupil dilation during a cognitive task), serves as one possible example. If someone has normal cognitive performance but is requiring greater mental effort to achieve that performance, they may experience that as cognitive decline, which could be a harbinger of later objective decline. ⁸ When assessing P-SCD or C-SCD, it is also worth considering cultural differences that may affect a person's views about the meaning of aging-related cognitive changes. Regardless, the current, widely used definition of SCD is linguistically problematic in that it commandeers a standard language expression, thereby precluding description of the personal experience of SCD in individuals who have also objectively declined. The proposed distinction between P-SCD and C-SCD provides a simple solution.