Cognitive Impairment, Dementia and Depression in Older Adults

1. Introduction

Cognitive impairment and depression are among the most prevalent neuropsychiatric conditions in late life. They frequently co-occur and exert reciprocal influences on each other [1,2]. Epidemiological studies have consistently shown that late-life depression is associated with a higher risk of developing mild cognitive impairment (MCI) and dementia, while cognitive decline often contributes to the emergence or persistence of depressive symptoms [3,4]. Both conditions share overlapping clinical features—such as deficits in attention, executive function, and processing speed—which can complicate differential diagnosis [5,6]. The concept of “pseudodementia,” once used to describe reversible cognitive impairment secondary to depression, has evolved into a more nuanced understanding of a potential prodromal stage of dementia, highlighting the blurred boundary between affective and neurodegenerative disorders in older adults [7].

Neurobiologically, depression and dementia exhibit several shared pathophysiological pathways, including dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis, neuroinflammation, cerebrovascular burden, and reduced neuroplasticity [8,9,10]. Late-onset depression, in particular, has been linked to frontostriatal and limbic system alterations as well as white matter hyperintensity (WMH) burden, suggesting a “vascular depression” subtype that bridges mood disturbance and cognitive decline [11,12]. Conversely, in established dementia, depressive symptoms are often part of the behavioral and psychological symptoms of dementia (BPSD), reflecting the spread of neurodegeneration into emotion-regulating networks [13,14,15]. These bidirectional and stage-dependent interactions underscore the need for an integrative framework that considers depression and cognitive decline as part of a continuum rather than distinct entities.

Understanding the interplay between mood and cognition has important therapeutic implications. Accumulating evidence suggests that antidepressant treatment may improve not only mood but also specific domains of cognitive performance, particularly when targeting serotonergic and neurotrophic mechanisms [16,17]. Similarly, while cognitive enhancers are known to modestly alleviate behavioral symptoms in dementia, recent advances in anti-amyloid and anti-tau antibody therapies have raised expectations that such disease-modifying treatments might also ameliorate affective symptoms, although definitive evidence is still lacking [18,19]. This narrative review aims to synthesize current evidence on the shared and distinct cognitive profiles of depression and dementia, explore the neurobiological and clinical features of depression as a prodromal or comorbid manifestation of dementia, and discuss the bidirectional treatment effects that bridge mood and cognition in late life. The review is based on a focused evaluation of epidemiological studies, neuroimaging and biomarker research, and major clinical trials relevant to late-life depression and dementia.

2. Shared and Distinct Patterns of Cognitive Impairment in Depression and Dementia

Cognitive impairment in late-life depression and dementia frequently involves deficits in attention, processing speed, and executive function, thereby creating a significant overlap in neuropsychological profiles. For example, older adults with major depressive disorder often demonstrate reductions in processing speed and executive functioning tasks, pointing to dysfunction within frontostriatal circuits [20,21]. Similarly, in the early stages of dementia, the decline in attention and processing speed may precede overt memory impairment, implicating frontal lobe involvement in both conditions [22,23]. This overlap poses a diagnostic challenge in distinguishing a reversible “pseudo-dementia” picture from a progressive neurodegenerative process, underscoring the need for nuanced neuropsychological assessment.

Despite these shared features, important distinctions in cognitive profiles emerge between depression and dementia. Memory dysfunction, particularly in encoding and storage of new information, is a hallmark of Alzheimer’s disease (AD) and other neurodegenerative dementias [24], whereas in depression the memory impairment tends to be more retrieval-based with recognition relatively spared [25]. In addition, cognitive deficits in depression may demonstrate substantial recovery following effective mood treatment, indicating a degree of reversibility; by contrast, cognitive decline in dementia is characteristically progressive and largely irreversible [7,26]. This divergence in trajectory and reversibility strengthens the conceptual distinction between functional cognitive impairment and early neurodegeneration.

From a neurobiological and neuroimaging perspective, depression and dementia share disruption in frontal–subcortical and limbic networks, yet they diverge in anatomical and pathophysiological patterns [27,28,29]. In late-life depression, evidence points to altered dorsolateral prefrontal cortex and anterior cingulate cortex functioning, increased WMH burden, and a “vascular depression” pattern that implicates cerebrovascular burden [27,29]. By contrast, dementia—particularly AD—exhibits prominent hippocampal and medial temporal-lobe atrophy, posterior cortical hypometabolism, and accumulation of amyloid and tau pathology [30]. Thus, while both conditions may involve frontolimbic network disruption, the underlying substrate in depression tends to be more vascular/functional, whereas in dementia it is more degenerative/structural.

Clinically, the recognition of this continuum from depression through cognitive impairment to dementia has considerable implications. Older adults presenting with late-onset depression, marked executive dysfunction, and high WMH burden are at increased risk of subsequent cognitive decline and conversion to dementia [31,32]. Accordingly, neuropsychological assessment in depressed older patients should incorporate not only mood-symptom resolution but also longitudinal cognitive monitoring, especially in the domains of processing speed and executive control [33]. Understanding both the similarities and differences in cognitive impairment between depression and dementia enables more precise stratification of risk and informs timely intervention strategies.

3. Depression as a Risk Factor for Dementia

Meta-analytic evidence strongly supports depression as an independent risk factor for subsequent dementia. In a comprehensive meta-analysis of 23 cohort studies, Diniz et al. (2013) reported that late-life depression was associated with an 85% increased risk of all-cause dementia (pooled hazard ratio [HR] 1.85, 95% confidence interval 1.67–2.04), with subtype analyses showing a stronger association for vascular dementia (odds ratio [OR] 2.52) than for AD (OR 1.65) [1]. Similar findings have been replicated in other large population-based cohorts and systematic reviews, consistently demonstrating that depressive disorders across the lifespan are linked to higher incidence of MCI and dementia [3,4]. More recent nationwide data from Denmark confirmed this association across early-, mid-, and late-life depression, showing HRs of 3.08, 2.95, and 2.31, respectively [34]. The impact of depression on dementia risk also appears to vary by age: a recent Korean cohort study of adults aged ≥ 75 years suggested that the association weakens with advancing age [35].

Several biological mechanisms have been proposed to explain the link between depression and later dementia. Among these, vascular pathology appears to play a particularly prominent role. Neuroimaging and neuropathological studies have consistently shown that older adults with depression often exhibit increased WMH, lacunar infarcts, and small-vessel disease, collectively forming the substrate of the so-called “vascular depression” subtype [11,27]. These cerebrovascular alterations may disrupt frontostriatal and limbic circuits that are essential for executive and emotional regulation, providing a plausible mechanistic bridge between mood disturbance and cognitive decline [9]. Conversely, while AD pathology—amyloid-β and tau accumulation—has been postulated to underlie the depression-dementia relationship, biological evidence for a direct association remains inconsistent. A recent systematic review and meta-analysis of cerebrospinal fluid, positron emission tomography (PET), and plasma studies found no overall association between amyloid-β burden and depression in older adults without dementia, although subgroup analyses suggested a possible link in cognitively impaired individuals [36]. Emerging evidence also implicates neuroinflammatory and oxidative mechanisms in the depression–dementia continuum, with elevated peripheral inflammatory markers such as interleukin-6 being linked to late-life depression [37], microglial activation and HPA-axis dysregulation contributing to neurodegenerative vulnerability [38], and oxidative stress biomarkers such as nitrotyrosine predicting subsequent dementia in depressed older adults [39]. Thus, while depression may coexist with Alzheimer’s pathology in some individuals, current evidence indicates that alternative biological pathways, rather than amyloid–tau accumulation alone, may contribute to the observed association with cognitive decline.

4. Depression as a Prodromal Symptom of Dementia

While depression has been clearly documented as a risk factor for dementia, accumulating longitudinal data suggest that in a subset of older adults, depressive symptoms may function as a prodromal manifestation of an incipient neurodegenerative process. For example, a large cohort study found that individuals with late-life depression exhibited a significantly elevated rate of dementia onset within the first few years of diagnosis, indicating temporal closeness to dementia onset rather than a lengthy latency period [40]. Moreover, systematic reviews have debated whether late-onset depression represents an early stage of AD or vascular dementia pathology rather than merely a comorbid mood disorder [41]. These observations highlight the importance of viewing certain episodes of late-life depression not just as antecedent risk factors, but as clinical signals of evolving brain pathology.

Although prodromal depression is difficult to distinguish clinically from non-prodromal forms of late-life depression, biomarker studies suggest that biological differences may emerge in specific subgroups. For example, Wu et al. (2016) [42] reported higher cortical amyloid burden in patients with late-life depression compared with healthy controls using ¹⁸F-florbetapir PET, although this association was evident only among those with co-occurring amnestic MCI. Late-onset depression has been associated with a higher burden of WMH and fronto-subcortical circuit disruption, which correlate with greater cognitive impairment [43]. Moreover, older adults with depression who exhibit substantial WMH burden tend to show poorer antidepressant treatment response, suggesting that cerebrovascular pathology contributes to both affective and cognitive outcomes in this population [11,27]. In one large study, the combination of late-life depression and subjective cognitive decline (SCD) was associated with substantially higher incidence of MCI and dementia compared with either condition alone [44]. These features suggest that depressive episodes occurring later in life may mask or overlap with very early neurodegenerative or vascular brain changes.

Recognizing depression as a potential prodromal symptom of dementia has important prognostic and clinical ramifications. Older adults with new-onset depression—especially those exhibiting executive dysfunction, SCD, and neuroimaging markers of cerebrovascular or hippocampal damage—appear to progress more rapidly to dementia than those without such features [45]. Accordingly, it is prudent for clinicians to adopt a protocol of longitudinal cognitive monitoring in these patients, extending beyond mood symptom remission to regular assessment of processing speed, executive control, and memory domains, as well as consideration of neuroimaging and vascular risk management. Early detection of the transition from depressive state to cognitive decline may facilitate timely intervention and possibly delay or mitigate progression to full-blown dementia.

5. Depression as a BPSD

Depressive symptoms are among the most common BPSD, affecting an estimated 30–50% of patients with AD and even higher proportions in dementia with Lewy bodies (DLB) and frontotemporal dementia (FTD) [14,15]. The prevalence of clinically significant depression is particularly elevated in DLB, where affective symptoms frequently coexist with fluctuations in attention and visual hallucinations [46], and in FTD—especially the behavioral variant—where apathy and mood changes are prominent early features [47]. The presence of depression in dementia is associated with faster cognitive and functional decline, greater neuropsychiatric burden, increased rates of institutionalization, and reduced quality of life for both patients and caregivers [48,49]. Several longitudinal studies further indicate that dementia patients with comorbid depression experience higher mortality and are more likely to exhibit severe agitation, psychosis, and loss of independence [50]. In addition, a few cohort studies have shown that the presence of depression in dementia is associated with accelerated cognitive decline, with one longitudinal analysis reporting that patients with all-cause dementia lost an average of 13.7 Mini-Mental State Examination scores over 3 years, and that comorbid depression contributed to additional cognitive loss beyond this baseline trajectory [51]. These epidemiological and clinical findings underscore that depression emerging in the context of established dementia is not a benign comorbidity but a major determinant of disease trajectory and patient outcomes.

In established dementia, depressive symptoms are thought to reflect neurodegenerative changes involving monoaminergic pathways, including degeneration of serotonergic and noradrenergic nuclei [52]; structural atrophy in limbic and paralimbic regions, such as the anterior cingulate cortex, hippocampus, and orbitofrontal cortex [53]; and disruption of the default-mode and salience networks [54]. Neuroinflammatory activation and cholinergic deficits may further contribute to affective symptoms, particularly in AD and DLB [55,56]. Treatment response is also distinct from depression in non-demented individuals. Large randomized trials have shown limited or modest efficacy of selective serotonin reuptake inhibitors (SSRIs) in AD [57], and symptom improvement is often heterogeneous and incomplete [58]. In contrast, cholinesterase inhibitors and multimodal non-pharmacological interventions, including caregiver training and structured activity programs, have been shown to confer modest benefits for depressive and apathy-related symptoms [59,60]. These mechanistic and therapeutic patterns highlight the need for dementia-specific approaches when managing depression as part of the broader BPSD syndrome. The distinguishing clinical and biological characteristics of late-life depression, prodromal depression, and BPSD-related depression are summarized in Table 1.

6. Reciprocal Effects of Antidepressant and Anti-Dementia Treatments on Cognition and Mood

Antidepressant treatment in late-life depression has been associated not only with mood improvement but also with subtle cognitive benefits in specific domains. Several clinical studies have shown that SSRIs may enhance processing speed, executive functioning, and working memory in older adults without dementia, particularly when depressive symptoms remit adequately [61,62]. Agents with multimodal serotonergic actions, such as vortioxetine, have demonstrated pro-cognitive effects independent of mood change in geriatric populations [63,64]. In contrast, randomized trials in AD have produced inconsistent findings. The HTA-SADD trial and other large controlled studies found no significant advantage of SSRIs over placebo for depressive symptoms or cognitive outcomes in AD, underscoring the limited cognitive efficacy of conventional antidepressants once neurodegeneration is established [57,65,66]. Thus, while mood treatment may produce modest cognitive gains in non-demented older adults, such effects diminish substantially in the context of progressive dementia.

Conversely, anti-dementia pharmacotherapies appear to exert modest benefits on depressive and apathy-related symptoms. Cholinesterase inhibitors, including donepezil, rivastigmine, and galantamine, have been reported to reduce neuropsychiatric symptoms in AD and related disorders, with several trials demonstrating small but measurable improvements in depression, irritability, and apathy [67,68]. Rivastigmine, in particular, has shown benefit in DLB, where enhancement of cholinergic transmission may alleviate mood symptoms alongside improvements in attention and behavioral fluctuations [69]. Memantine, an N-methyl-D-aspartate receptor antagonist, has also been associated with reductions in agitation, irritability, and occasionally depressive symptoms, although results remain heterogeneous across studies [70]. Overall, although the magnitude of these effects is modest, the evidence suggests that targeting cholinergic and glutamatergic dysfunction in dementia may have secondary benefits on mood and behavioral outcomes.

Emerging disease-modifying treatments further highlight the complex interplay between cognition and affect in neurodegenerative disorders. Anti-amyloid monoclonal antibodies such as lecanemab [71] and donanemab [72] have demonstrated clinically meaningful slowing of cognitive decline in early Alzheimer’s disease; however, current trial data do not provide clear evidence of improvement in depressive symptoms, and any potential mood effects remain speculative [73,74]. Similar uncertainty surrounds tau-targeting therapies, which are primarily designed to modify neurodegenerative progression rather than affective symptoms [75]. Nevertheless, the possibility that reducing pathogenic burden could indirectly modulate affective circuits has prompted interest in mood-related exploratory outcomes in ongoing trials [73]. Additionally, multimodal approaches—including cognitive rehabilitation, structured exercise programs, caregiver-focused interventions, and neuromodulation techniques such as repetitive transcranial magnetic stimulation—show promise for dual benefits on mood and cognition [76,77,78]. Together, these findings point to an emerging therapeutic paradigm in which interventions increasingly target the intertwined trajectories of cognitive decline and affective disturbance in late life. A comparative overview of the cognitive and mood-related effects of antidepressants, cognitive enhancers, and emerging disease-modifying therapies is provided in Table 2.

7. Conclusion

Depression and cognitive impairment in late life are deeply interconnected. They share overlapping clinical manifestations and partially convergent neurobiological pathways. Evidence across epidemiological, neuroimaging, and biomarker studies demonstrates that depression can function both as an independent risk factor for dementia and, in some individuals, as a prodromal expression of emerging neurodegenerative or vascular pathology. At the same time, a substantial proportion of patients with established dementia experience depressive symptoms as part of the broader BPSD spectrum, where mood changes further accelerate functional decline, worsen neuropsychiatric burden, and negatively affect patient and caregiver quality of life. These stage-dependent relationships underscore the importance of moving beyond a binary distinction between “depression” and “dementia,” and instead adopting an integrative conceptual framework that captures their dynamic and bidirectional interactions.

Therapeutically, the reciprocal effects of mood- and cognition-targeted treatments highlight both opportunities and limitations. Antidepressants may yield modest cognitive benefits in non-demented older adults, but their efficacy is substantially reduced in dementia, where cholinesterase inhibitors, memantine, and multimodal non-pharmacological interventions provide small but measurable improvements in affective and behavioral symptoms. Meanwhile, emerging disease-modifying therapies offer hope for altering the trajectory of Alzheimer’s pathology, though definitive evidence for mood effects is lacking. These findings emphasize the need for personalized, stage-specific assessment and management strategies, incorporating longitudinal cognitive monitoring, vascular and neuropsychiatric risk evaluation, and multimodal interventions tailored to the evolving clinical profiles of aging individuals.