Older Adults’ Self-Reported Prospective Memory Lapses in Everyday Life: Connections to Inflammation and Gender

Introduction

Extensive research efforts have been dedicated to improving early identification of symptomology and risk for cognitive decline and Alzheimer’s disease and related dementias (ADRD). Inflammatory responses appear to play an important role in Alzheimer’s disease (AD)-specific pathology and blood-based inflammatory markers may offer opportunities for early identification of risk without requiring more time-consuming or invasive procedures (e.g., brain imaging, cerebral spinal fluid analysis). Blood-based inflammatory markers implicated in risk for cognitive decline and ADRD include pro-inflammatory cytokines (e.g., interleukin [IL]-1β, IL-6, IL-8, tumor necrosis factor [TNF]-α), anti-inflammatory cytokines (e.g., IL-4, IL-10), and C-reactive protein (CRP)[1–7]. Much of the work connecting inflammation to risk for cognitive decline has examined diagnostic or laboratory-based cognitive performance; less is therefore known about how inflammation relates to individuals’ everyday cognitive functioning. An emerging body of work has partially addressed this knowledge gap by focusing on links between inflammation and subjective cognitive reports[8–10].

Subjective cognitive reports related to more frequent forgetting may be one of the earliest symptoms of AD[11] and should serve as a red flag for clinicians. Although ADRD risk appears to be even stronger when subjective reports of forgetting are linked with biological risk factors[12], limited research has examined the links between inflammation and subjective cognition. Findings in this area have been mixed, with some pro-inflammatory cytokines (i.e., monocyte chemotactic protein-1) correlating with greater subjective cognitive complaints, and others (i.e., TNF-α) correlating with fewer complaints[10]. Other work that examined differences in gene expression involved in certain inflammatory responses between individuals with and without subjective memory complaints found no significant differences between groups[9]; however, this study and others like it use brief single-time point assessments that broadly address subjective cognition, which may not be sensitive enough to capture daily experiences. Rather than general subjective assessments, examining specific types of forgetting in daily life may improve our ability to speak to unique links between inflammation and cognitive risk factors for future decline and pathology. Accordingly, the present study focuses on the links between inflammatory markers and subjective reports of prospective memory lapses in daily life.

Prospective memory (PM) refers to our memory for future actions or events[13], such as remembering to turn off the stove after cooking or attending a scheduled appointment. Although there are some mixed findings[14], individuals are more likely to experience PM lapses in daily life compared to other memory lapses, such as retrospective memory failures[15,16]. Much of the existing PM research documents task-specific[17] or psychosocial factors[18] related to older adults’ PM performance. Yet, person-specific predictors, including biological factors like inflammation, may play a critical role in PM performance and decline and require further attention. PM forgetting can be consequential for maintaining independence into older adulthood[19,20] and may help distinguish healthy older adults from those in the early stages of ADRD earlier than deficits in other cognitive abilities like executive function, retrospective memory, and working memory[21,22]. Despite arguments that PM decline may be an early indicator of pathological memory aging, limited work has addressed PM in relation to biological correlates of cognitive decline and ADRD risk[22–25].

To our knowledge, only one study has directly assessed the link between inflammation and laboratory-based PM[26]. In this study, researchers examined circulating inflammatory cytokines (i.e., IL-1β, IL-1 receptor antagonist, soluble IL-1 receptor type-II, soluble IL-2 receptor, and IL-6) within a sample of women (aged 45–90 years) who were asked to remind the experimenter at the end of the testing session to sign a paper. Higher levels of IL-6 related to poorer PM performance only among older women (aged 60+). However, the generalizability of this finding is unclear given that older adults’ PM performance often differs between laboratory and naturalistic settings[27] and single-item assessments of PM often lack reliability[28,29]; moreover, this study only included women.

There is reason to believe that the link between PM and inflammation would differ by gender/sex¹. Women are more likely than men to experience other conditions associated with inflammation (e.g., pregnancy complications, depression, chronic stress) which are also linked with cognitive impairment, including poor PM[32,33], and increased risk for ADRD later in life[34–36]. Conversely, men are at greater risk for vascular cognitive impairment across the lifespan compared to women[34], with heightened inflammation involved in the underlying pathology for this greater risk. Researchers have noted a dearth of studies that focus on gender differences in ADRD risk as well as the importance of identifying sexual dimorphisms to improve early detection, prevention, and care for those with cognitive decline and ADRD[36,37]. To the extent that PM forgetting is an early ADRD indicator[21,22], identifying gender differences in the link between inflammation and subjective PM reports may be critical in determining who is at greater risk of daily memory failures and potential future pathology.

Method

Results

Participants reported experiencing 0 to 35 PM lapses across the two-week measurement period (M = 5.48, SD = 6.74). Approximately 24% of our sample reported zero PM lapses across the two-week measurement period. When examining the relationships between PM lapses and the inflammatory markers within the sample as a whole (Table 2), the only significant link to emerge was in relation to basal IL-8, such that a greater number of reported PM lapses was correlated with higher basal levels of IL-8 (r = .15, p = .024). However, this relationship was not robust to FDR adjustments (adjusted p = .468); thus, no follow-up ZINB regression analyses were conducted. When analyses were stratified by gender/sex, men’s basal IL-8 correlated with PM lapses, such that higher levels of basal IL-8 were associated with more PM lapses (r = .40, p < .001). This relationship remained significant after FDR adjustments (adjusted p = .039). Women’s reported PM lapses did not relate to basal IL-8 (r = .02, p = .775). No other significant relationships between inflammatory markers and PM lapses emerged among men or women.

Following-up the observed gender difference in the relationship between basal IL-8 and PM lapses, we conducted gender stratified ZINB analyses³ (Table 3). Among men, analyses revealed a significant effect of basal IL-8, such that higher levels of basal IL-8 were associated with reporting more PM lapses (b = 0.98, p < .001), thereby supporting of our initial correlational findings. Again, ZINB analyses indicated this relationship was not significant among women (b = 0.03, p = .826).

On an exploratory basis, we examined gender/sex differences in the types of reported PM lapses (see Supplemental Materials)⁴. In brief, more men (34.2%) than women (18.4%) reported forgetting to complete an errand or chore (p = .008), whereas more women (42.1%) than men (21.5%) reported forgetting to complete an action they had previously started at least once across the two-weeks (p = .002). Among men, higher basal IL-8 related to greater instances of forgetting to complete an errand/chore, take medication, make a phone call, and bring something (ps < .05). Alternatively, men’s higher basal IL-1β and IL-6 related to fewer instances of forgetting to complete a previously started action (ps < .05). Among women, basal IL-8, again, did not relate to PM lapses. However, higher levels of basal IL-1β and IL-4 in women related to a greater instance of forgetting to attend a meeting/appointment (ps < .05). Additionally, higher levels of CRP and several stimulated cytokines (i.e., IL-6, IL-8, IL-10, TNF-α) correlated with a lower likelihood of certain types PM lapses among women (ps < .05), mostly with forgetting to attend a meeting. As these analyses are exploratory, findings should be interpreted with caution.

Discussion

The current study examined relationships between subjective reports of PM lapses in everyday life and inflammatory markers within a diverse sample of older men and women. Contrary to hypotheses, self-reported PM lapses did not relate to most inflammatory markers within our sample as a whole. Although higher levels of basal IL-8 related to greater reported PM lapses within our sample as whole, this link was not robust to FDR adjustments. These null findings are, in part, supported by work from Lekander and colleagues[26] who did not observe a relationship between PM and IL-1β, and from Baune and colleagues[1] who found a significant association between memory performance and basal IL-8, but not with IL-1β, IL-4, IL-6, or IL-10. As only one known study has examined inflammatory cytokines in association with PM[26], we decided to include all available inflammatory indicators to better discern the relationship between self-reported PM lapses and inflammation. It is possible that older adults’ self-reported PM is not strongly associated with inflammation, at least not at cross-section. Continued investigation into the longitudinal trajectory of inflammatory markers and PM lapses may reveal additional associations that the current cross-sectional study was unable to identify.

A second key goal of the present work was to examine our hypothesis that associations between inflammatory markers and PM lapses would differ by gender. We found that higher levels of basal IL-8 were linked with reporting more PM lapses only in men, an association that was robust to FDR adjustments and covariates. This finding is novel, as the only known study to have examined a connection between inflammation and PM lapses did not measure IL-8 and included only women[26]. In related work (albeit not specific to PM lapses), higher levels of IL-8 related to poor memory performance among older adults[1,50], but gender/sex differences were not examined. Another study that examined the relationship between IL-8 and memory performance[51] did not observe gender/sex differences; however, this study utilized neurocognitive assessments that did not include a measure of PM and furthermore did not capture cognition in everyday settings. More related to cognitive reports, Iulita and colleagues[52] found that individual with subjective cognitive impairment had slightly higher levels of IL-8 compared to healthy controls and that higher levels of IL-8 correlated with cognitive decline; unfortunately, this past study was underpowered to address gender differences in the observed links. The present work builds upon these past studies and suggests that self-reported PM forgetting in daily life relate to IL-8 only among older men.

There are several possible explanations as to why PM lapses were uniquely linked with levels of IL-8 in the present research. Researchers who observed similar relationships between higher IL-8 and poor cognitive functioning argued that this link may be indicative of the role of IL-8 in neurodegeneration[1]. IL-8 is a pro-inflammatory chemokine that is involved with early signaling in relation to damage[53], and is thought to mediate glial interactions with neurons and contribute to neuronal damage[54]. Higher levels of IL-8 have been associated with white matter hyperintensity volume (WMHV); WMHV has been associated with reduced cognitive performance in men[55], though this effect is not confined to men[56]. Taken together with recent research that documents poorer PM performance among individuals with more severe WMHs[57], PM lapses may occur more frequently in men with elevated IL-8 due to white matter damage. Furthermore, elevated levels of IL-8 have been found within circulating blood[58,59], cerebral spinal fluid[58], and postmortem brains of patients with AD[60]. Importantly, an upregulation of IL-8 has been suggested to occur early in the progression of AD[60] and can be considered an early risk factor for ADRD[61]. As researchers have argued that PM decline may be an early indicator of dementia[21,22], the observed elevated basal IL-8 levels in older men may be an early indicator of neurodegeneration that impacts everyday PM. Replication of the current findings and additional work is needed to discern whether this association is relevant to underlying neuropathology and/or ADRD risk in men.

Although we hypothesized that the link between inflammation and PM would differ by gender/sex on an a priori basis, the specific pattern observed in the present research was not predicted. Additional research is needed to determine the generalizability of this finding and potential related mechanisms. From a biological perspective, differences in sex hormones may partially explain the gender/sex-specific relationship between basal IL-8 and PM. Women’s higher levels of estrogen may protect aspects of cognition from the detrimental effects of inflammation[62], but research has yet to examine gender/sex differences in PM in relation to sex hormones. It is also important to recognize that there may be societal determinants, such as the division of mnemonic responsibilities, that influence the gendered effects of inflammation on PM. Women are often expected to help their male partners[63,64] and children remember their PM intentions[65]. In doing so, women may develop better mechanisms (e.g., strategy use) to manage the cognitive demands related to retaining familial PM intentions[65,66] which, in turn, might buffer PM from the effects of inflammation. If men rely on women to assume responsibility for or assist with their PM tasks, they may not develop the necessary tools to support their memory, making them more susceptible to lapses, especially in the presence of elevated IL-8.

Regarding links among women, we did not replicate Lekander and colleagues’ finding of an association between older women’s PM and IL-6 levels[26]. The discrepancy between findings may stem from differences in PM measurement (i.e., laboratory-based vs self-report). PM researchers have long observed differences between older adults’ performance on laboratory and naturalistic PM tasks[27,67]. Furthermore, self-generated PM tasks, like those assessed in this study, can be more susceptible to interference from daily life and are arguably more representative of a person’s PM abilities than experimenter-generated tasks[38,67]. Hence, associations between PM and inflammation might differ across testing settings or across task types (e.g., experimenter- or self-generated). Rather than focusing on women’s number of lapses, the present exploratory analyses suggest that examining types of PM lapses in relation to inflammation may be more informative for understand women’ everyday PM; yet, additional research is required to replicate these exploratory findings. Longitudinal examinations may also reveal links between women’s PM lapses and inflammation that we were unable to identify. Past work suggests that elevated levels of CRP predict poor memory performance among women during a 12-year follow-up[68]; importantly, the composite memory performance score in this study included PM. It is thus possible that PM itself may relate to inflammation across time, but this notion is currently untested. Given women’s higher ADRD risk, it is particularly important to identify risk factors or early indicators of pathology among this group[37]. Continued research into inflammatory markers and cognition, including PM, across a variety of tasks and settings will help elucidate underlying causes of gender/sex differences in cognitive decline and dementia risk.

Supplementary Material