Introduction
Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by pathological aggregation of the proteins amyloid-β (Aβ) and tau in the brain [
52]. As Aβ plaques and neurofibrillary tau tangles form, communication between neurons is disrupted leading to atrophy, and ultimately functional impairment affecting multiple cognition domains (e.g., memory, visuospatial ability, language, and attention). Although debilitative functional changes occur with the progression of disease, it has been suggested that the pathophysiology of AD begins nearly 20 years prior to the clinical presentation of symptoms [
54,
61]. Therefore, it has become critical to target AD-related biomarkers early as they may be reflective of future decline. Advancements in clinical trial research have resulted in the US Food and Drug Administration’s (FDA) recent approval of Aducanumab (Aduhelm) for AD treatment. However, due to its lack of effectiveness at improving cognitive functioning, conflicting trial results, and potential harm caused by the drug [
27,
37,
55], agencies such as the European Medicines Agency (EMA) have refused to market this medication as a treatment for AD. As such, to date, there is still no uniform drug or treatment available to slow the progression of disease or reverse the disease process. For this reason, it has become increasingly important for research targeting mechanisms of early detection to help with disease prevention and to combat the deleterious effects of AD.
Research has suggested that subjective cognitive decline (SCD), or the self-reported experience of subtle changes in cognitive functioning without any measurable changes in neuropsychological test performance [
25], may be a preclinical marker of AD [
3]. For example, individuals who endorse SCD have an increased risk of developing AD compared to the general population [
3,
49,
53]. Typically reported as increased confusion or memory loss, the prevalence of SCD among adults aged 60 and older is around 25% [
50]. Therefore, SCD may prove to be an effective target for early intervention. Existing intervention studies and clinical trials targeting AD and other related dementia risk factors have revealed limited success at maintaining or improving cognitive function [
1,
17,
60]. However, these interventions are often introduced after AD-related cognitive decline has already begun. This critical time window may explain why such interventions have difficulties in demonstrating effective ways of preventing and reducing cognitive decline. Although interventions have been largely unsuccessful in terms of elimination of symptoms, it has been suggested that they may be effective at ameliorating symptoms, thus improving cognitive outcomes [
41]. SCD has been suggested as one of the earliest clinical indicators of AD prior to measurable cognitive decline, and its cognitive correlates align with the earliest pathological changes in AD. Therefore, explorations into SCD may allow for the improvement of early detection techniques at a critical time window prior to more pronounced atrophy and objective clinical symptoms as a consequence of disease progression.
Sex differences have also been observed in SCD, albeit findings yield inconsistent results. For example, it has been revealed that SCD in females is more strongly associated with future dementia diagnoses than in males [
19]. Another study has observed that SCD in males is associated with worse performance on a measure of global cognition (Alzheimer’s Disease Assessment Scale-13) compared to females [
56]. While the former suggests that SCD is associated with clinical progression in females more strongly than males, the latter indicates that SCD is associated with increased cognitive decline in males compared to females. Taken together, findings from these studies reveal that biological sex may play a role in SCD. However, further research is needed to better understand the interaction between sex and SCD. Despite evidence indicating biological sex is independently associated with AD prevalence [
12,
45], cortical atrophy [
15,
29], and clinical progression [
14,
62], the relationship between sex, cognition, and SCD classification remains relatively unexplored. It is critical to investigate whether cognitive decline observed in people with SCD differs as a function of biological sex. Such an exploration may result in a better understanding of whether females endorsing SCD have different cognitive trajectories subjecting them to greater decline and higher prevalence of AD compared to males. To examine this relationship, we investigated sex differences in cognition in a sample of female and male healthy older adults with and without SCD. The importance of examining sex differences in both groups is to ensure that the change over time is specific to those with SCD and not simply what occurs in healthy “normal” aging in this sample. This design allows us to determine whether SCD is predictive of future cognitive decline and if this association differs by biological sex.
Discussion
Previous findings have suggested that cognitive functioning, including rate of cognitive decline, may differ between males and females (e.g., [
32]. However, there is limited research examining sex differences in people who may be at the earliest stages of cognitive decline, those with SCD. This limited understanding of how sex may influence cognitive decline in SCD limits the ability to have targeted interventions and therapies to help prevent cognitive decline due to MCI or dementia. Therefore, the current study aimed to elucidate sex disparities in the trajectory of cognitive decline to aid in a better understanding of techniques for early detection and disease mitigation. In our sample of 3019 cognitively unimpaired older adults, the rate of change in cognitive performance varied between males and females in people with SCD. Specifically, males exhibited significantly lower baseline performance in global cognition, episodic memory, and perceptual speed but higher performance in visuospatial abilities. The three-way interaction between Sex, SCD classification, and Time From Baseline was also significant, revealing that SCD + females decline at a significantly faster rate than SCD + males in all cognitive domains. When examining the interaction effect of SCD classification and Time From Baseline in the longitudinal model, SCD + males exhibited significantly lower overall performance in global cognition, episodic, and semantic memory compared to SCD − males, while SCD + females exhibited significantly lower performance in all cognitive domains compared to SCD − females. Our results reveal that (1) people with SCD have both lower baseline cognition and an increased rate of decline compared to people without SCD, and (2) SCD in females may be more predictive of future cognitive decline than in males, which may help explain sex disparities in cognitive decline.
There is mounting evidence suggesting that sex differences exist in both normal cognition and dementia. For example, a recent cohort study of over 26,000 participants reported cognitively unimpaired females to have greater global cognition, executive function, and memory compared to males [
32]. Similarly, in our study, a main effect of sex was observed for all cognitive domains except working memory, demonstrating that regardless of SCD status, females tend to score higher on neuropsychological assessments in several domains compared to males, whereas males score higher in visuospatial ability then females. Levine and colleagues [
32] also observed that cognitively unimpaired females had an increased rate of decline compared to males. Additionally, several other studies have suggested that although females may score higher at baseline, they may be subject to faster cognitive decline compared to males [
21,
23,
34]. This increased rate of decline in females may contribute to the sex disparities that exist in prevalence of AD [
8,
35]. In the present study, females were not observed to have an increased rate of decline in the SCD − group compared to males. Rather, sex differences in cognitive decline were observed only in SCD + group. SCD + females experienced steeper declines in cognitive performance compared to SCD + males, and SCD − males and females, in all domains. That is, although SCD − females are shown to consistently have the highest cognitive performance over time compared to SCD − males (except in visuospatial abilities), and SCD + males and females, the introduction of SCD ( +), negatively affects this relationship, which may be indicative of a more rapid trajectory of cognitive decline.
While previous research has examined sex differences in SCD status, the results are limited to smaller samples and cross-sectional data [
56] as well as only examining subsequent dementia (and AD) conversion [
19,
47], global cognition, and instrumental activities of daily living [
47] and not rate of cognitive change. Previously, females with SCD have been reported to exhibit lower overall baseline global cognition, enhanced memory, and similar performance on executive functioning and semantic memory tasks compared to males with SCD [
56]. The current study expanded on this work by examining a large sample with longitudinal assessments of cognitive performance in order to further explore sex discrepancies. The longitudinal assessment is particularly important given that at baseline, females exhibited enhanced performance than males in most domains (except semantic memory and visuospatial ability), whereas females with SCD + had an increased rate of decline in all domains (i.e., global cognition, episodic memory, semantic memory, perceptual speed, visuospatial ability, and working memory). Furthermore, in the current study, we focused on examining sex differences in the trajectory of cognitive decline over time, rather than the relationship between worry and conversion in individuals with SCD. Future research should examine whether sex differences are present in the relationship between SCD and worry with cognitive decline.
Given that SCD (SCD +) has been linked to increased risk for MCI and AD [
24], those who endorse these subjective cognitive complaints are at greater risk for subsequent cognitive decline compared to those that do not (e.g., [
44]). In the current study, both male and female SCD + participants had increased rate of decline compared to SCD − participants, supporting the notion that SCD is indicative of future decline [
24,
28]. However, SCD + males only had increased rates of decline in global cognition, episodic memory, and semantic memory compared to SCD − males, whereas SCD + females decline at significantly faster rates than SCD − females in all cognitive domains. When employing the more conservative approach, SCD + males did not have increased rates of decline in any domain compared to SCD − males. With the presence of sex differences in SCD + (i.e., SCD + females exhibiting increased rates of change in all cognitive domains compared to SCD + males), our findings suggest that the relationship between SCD and future cognitive changes may be more predictive of cognitive decline in females compared to males. That is, females reporting SCD may be more likely to experience substantial cognitive changes compared to males with SCD. These findings suggest that waning cognitive abilities may have the potential to be captured early, particularly in females, with SCD + individuals detecting subtle cognitive changes prior to objective testing. Previous studies have observed that females tend to self-report cognitive changes more than males [
38]. Combining our findings with the increased reports in females relative to males may be indicative of either greater changes or better perception of cognitive changes in females. As such, the relationship between SCD and cognition may be stronger in females.
Findings from this study have important implications for interventions and therapies designed to target cognitive decline and dementia prevention. For example, risk factors such as midlife hypertension, midlife obesity, diabetes, physical inactivity, smoking, depression, and low education are all modifiable factors contributing to 1/3
rd of all AD cases [
46]. However, several of these AD risk factors disproportionately affect females. For example, both lower educational attainment, as well as psychiatric disorders such as depression, are more prevalent in females [
22]. Additionally, blood pressure is observed to be higher in males early in life, whereas females have a steeper increase in blood pressure that continues throughout the life compared to males [
26]. This prevalence of higher mid-life blood pressure in females is associated with a greater risk for the development of dementia compared to males [
11]. Other factors specific to females such as preeclampsia, menopause, and hypertensive pregnancy disorders also have negative impacts on the cardiovascular system and cognition [
16,
40]. These risk factors, paired with explanations such as higher life expectancy [
18], lower cognitive reserve, and faster rates of functional and structural deterioration [
31] in females compared to males have all led previous literature to reveal female sex to be a significant risk factor for AD [
8,
35]. The current study supplements the existing literature by revealing that in the earliest potential stage of the AD-trajectory prior to measurable cognitive decline (i.e., preclinical-AD or SCD), females also exhibit steeper declines in all cognitive domains over time compared to males. As such, directing therapies and interventions toward risk factors that have increased incidence in females may help reduce the prevalence of dementia in these individuals. Our findings suggest that SCD may be a critical indicator of subsequent cognitive decline in females, and therapeutic interventions may wish to target this population to better elucidate sex disparities in cognitive change over time.
One limitation of the current work is the use of only two questions to determine SCD status. Previous research has shown that different questionnaires used to determine SCD status results in different cognitive trajectories and atrophy patterns [
42] as well as different patterns of white matter hyperintensity burden [
43]. Therefore, it is thus possible that the use of different questionnaires may target specific declines in males vs. females and improve the relationship between SCD and cognition in males. Future research should explore this relationship. Females also have increased risk factors that influence vascular components. The resulting pathological changes due to vascular damage, such as white matter hyperintensities which are known to be associated with cognitive decline and conversion to dementia [
13], may be higher in females with SCD. Future research should examine the association between sex and SCD status on atrophy and white matter hyperintensities. Another limitation is that there are several factors that may influence the endorsement of SCD (e.g., depression, worry, and personality). Several studies have demonstrated that females are much more likely than males to experience depression [
48] and develop anxiety disorders due to excessive worrying [
36], and they tend to have distinctly different personalities compared to males [
57]. The fact that sex differences are prevalent in these domains affecting SCD may potentially contribute to disparities in the experience and subsequent endorsement of SCD. Although findings are inconsistent on the potential effects of these factors, they may play a role in either contributing to, or exacerbating, the association of SCD with the risk of decline and/or dementia [
19,
33]. Future research should further examine the influence of various covariates on SCD to determine its biological basis and whether these factors mediate the relationship between SCD and cognitive decline.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.