# paper.pdf ## Metadata - PDFFormatVersion=1.6 - Language=en - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260624185859+05'30' - Creator=LaTeX with hyperref package - ModDate=D:20260624185919+05'30' - Producer=Acrobat Distiller 26.0 (Windows) - Title=Associations of self‐reported obstructive sleep apnea with cognition and dementia risk in cognitively unimpaired middle‐aged adults - Custom.WPS-ARTICLEDOI=10.1002/alz.71553 - Custom.WPS-JOURNALDOI=10.1002/(ISSN)1552-5279 - Custom.WPS-PROCLEVEL=3 - xmp:createdate=2015-08-18T08:05:36+05:30 - xmp:modifydate=2016-06-21T14:17:29-04:00 - xmp:metadatadate=2016-06-21T14:17:29-04:00 - pdf:producer=Acrobat Distiller 9.0 (Windows) - pdfx:wps-articledoi=10.1002/alz.71553 - pdfx:wps-journaldoi=10.1002/(ISSN)1552-5279 - pdfx:wps-proclevel=3 - jav:journal_article_version=VoR - prism:doi=10.1002/alz.71553 - dc:format=application/pdf - dc:title=Associations of self‐reported obstructive sleep apnea with cognition and dementia risk in cognitively unimpaired middle‐aged adults - xmpmm:documentid=uuid:14490651-7765-3109-9776-220163ee6b2e - xmpmm:instanceid=uuid:524ff680-9f0e-3d6a-aa94-6df70fc03bc6 ## Contents ### Page 1 Received: 20 October 2025 Revised: 28 April 2026 Accepted: 29 April 2026 DOI: 10.1002/alz.71553R E S E A R C H A R T I C L E Associations of self-reported obstructive sleep apnea with cognition and dementia risk in cognitively unimpaired middle-aged adults Gabriel T. Abdelmessih Lisa Bransby Hannah Cummins Melinda L. Jackson Yen Ying LimTurner Institute for Brain and Mental Health, School of Psychological Sciences, Monash University, Clayton, Victoria, AustraliaCorrespondenceYen Ying Lim, 18 Innovation Walk, Monash University, Clayton, VIC 3800, Australia. Email: yenying.lim@monash.eduFunding informationNational Health and Medical Research Council, Grant/Award Numbers: GNT1158384, GNT1147465, GNT1111603, GNT1105576, GNT1104273, GNT1171816; Alzheimer’s Association; Dementia Australia Research Foundation; Bethlehem Griffiths Research Foundation; Yulgilbar Alzheimer’s Research Program; National Heart Foundation of Australia, Grant/Award Number: 102052; Australian Government Research Training Program (RTP) ScholarshipAbstract INTRODUCTION: Obstructive sleep apnea (OSA) is a potential risk factor for cognitive impairment and dementia; however, its contribution in midlife and interactions withAPOE ε4 remain unclear.METHODS: Participants were 2795 cognitively unimpaired, middle-aged adults enrolled in the Healthy Brain Project. OSA status was determined by self-report. Cog- nition was assessed using the Cogstate Brief Battery, and dementia risk using the Cardiovascular Risk Factors, Ageing, and Incidence of Dementia (CAIDE) score.RESULTS: Participants with OSA demonstrated poorer memory than those without OSA, although this association was attenuated after adjusting for vascular risk. Indi- viduals with OSA (with or without APOE ε4) had significantly higher CAIDE scores than those with neither risk factor. APOE ε4 did not moderate OSA-cognition associations.DISCUSSION: OSA may be associated with poorer memory and greater dementia risk, irrespective of APOE ε4 carriage. These findings highlight the need for early OSA screening to identify individuals at elevated dementia risk.K E Y W O R D SAlzheimer’s, APOE, cognition, dementia, memory, midlife, neuropsychology, OSA, sleep, vascular riskHighlights∙ OSA was associated with poorer memory in at-risk middle-aged adults, though this association was attenuated after adjustment for vascular risk factors.∙ OSA was also associated with elevated dementia risk, including risk driven by mod- ifiable vascular factors such as obesity, hypertension, hypercholesterolemia, and physical inactivity.∙ APOE ε4 did not moderate the associations of OSA with cognition or dementia risk. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2026 The Author(s). Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association.Alzheimer’s Dement. 2026;22:e71553. wileyonlinelibrary.com/journal/alz 1 of 12https://doi.org/10.1002/alz.71553 ### Page 2 2 of 12 ABDELMESSIH ET AL.∙ This work highlights that routine screening for OSA in midlife may help identify individuals at elevated dementia risk.1 BACKGROUNDObstructive sleep apnea (OSA) is a common sleep disorder charac- terized by recurrent collapse of the pharyngeal airway during sleep, causing intermittent hypoxemia and sleep fragmentation.1 OSA has emerged as a potential risk factor for late-life cognitive impair- ment and dementia, including Alzheimer’s disease (AD) dementia.2–5 Pathophysiological models propose that intermittent hypoxemia and sleep fragmentation directly contribute to oxidative stress, cere- brovascular injury, and impaired glymphatic clearance of amyloid-β, ultimately resulting in neuronal dysfunction and death.6 OSA may also indirectly result in cognitive impairment and dementia risk through its association with other established dementia risk factors, including hypertension,7 hypercholesterolemia,8 obesity,9 and physical inactivity.10 Importantly however, the relationship between OSA and the risk it poses on cognitive impairment and dementia remains equivocal, likely resulting from inconsistent findings to date. While studies in clinical populations more consistently support a relationship between OSA, cognitive impairment, and dementia risk,11–13 findings from community-based samples have been mixed. Some studies have found OSA to be associated with deficits in multiple cognitive domains and increased dementia risk,3,14–19 whereas others have reported no such relationships.20–23 These discrepancies may reflect methodological differences, including sample characteristics (e.g., clinic-based samples including individuals with more severe OSA), inadequate control for potential confounders (e.g., vascular risk factors), and the use of neu- ropsychological tests that lack sensitivity to early insults to the central nervous system. Thus, further research is warranted to clarify whether OSA is associated with cognition and dementia risk, particularly within community-based samples. An additional consideration in the relationship between OSA and cognitive impairment is the presence of the apolipoprotein E (APOE)ε4 allele, the strongest known genetic risk factor for sporadic AD.24 In older adults, carriage of even one copy of the APOE ε4 allele is associated with greater susceptibility to cerebrovascular disease and increased amyloid-β deposition.25–30 Emerging research suggests that in middle-aged and older ε4 carriers, OSA is more strongly related to poorer cognitive function than in non-carriers.14,31–35 However, the additive or synergistic influence of OSA and APOE ε4 on cognitive function and dementia risk remain unclear, particularly in middle-aged adults (e.g., 40–70 years). Investigating these relationships in midlife, before the onset of frank cognitive impairment or significant neu- ropathological changes, is crucial, as this is when dementia prevention strategies may be most efficacious.36 The first aim of this study was to investigate the associations of OSA with cognition and dementia risk, operationalized using the Cardio- vascular Risk Factors, Ageing and Incidence of Dementia (CAIDE) risk score, in a large, community-based sample of cognitively unimpaired, middle-aged adults enriched for a family history of dementia. It was hypothesized that individuals with OSA would exhibit poorer cognition and greater dementia risk compared to those without OSA. The second aim was to explore the nature and extent to which APOE ε4 carriage modified these associations.2 METHODS 2.1 ParticipantsThis cross-sectional study included baseline data for 2795 participants enrolled in the Healthy Brain Project (HBP; healthybrainproject.org.au; Figure 1). The HBP is a prospective, community-based online cohort of cognitively unimpaired adults, enriched with a family history of dementia.37 The recruitment and enrolment procedures for the HBP have been comprehensively detailed previously.37 Briefly, participants were required to be aged 40–70 years, reside in Australia, and be flu- ent in English to complete the online assessments. Exclusion criteria included self-reported cognitive impairment, neurological conditions (e.g., acquired brain injury, multiple sclerosis), neurodegenerative dis- eases (e.g., dementia, Parkinson’s disease), or any use of Therapeutic Goods Administration-approved medications for AD (e.g., donepezil, memantine). The HBP was approved by the Monash University Human Research Ethics Committee, and all participants provided informed consent via the online platform before study commencement. Data were col- lected between November 2016 and December 2024, with the current study including data collected up until the fifth data freeze (Decem- ber 2022). This report adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.382.2 Demographic characteristicsParticipants self-reported their date of birth, sex, height, and weight (used to calculate body mass index [BMI]), years of education, eth- nicity, and medical history (including history of hypertension, hyperc- holesterolemia, and sleep disorders). Participants also completed the Hospital Anxiety and Depression Scale (HADS),39 to determine depres- sive and anxiety symptoms; the Epworth Sleepiness Scale (ESS),40 to ### Page 3 ABDELMESSIH ET AL. 3 of 12RESEARCH IN CONTEXT1. Systematic review: The authors reviewed the literature using traditional sources (e.g., PubMed). While obstruc- tive sleep apnea (OSA) OSA has been associated with cog- nitive impairment and dementia, findings in community- based samples have been inconsistent. Evidence also suggests that OSA interacts with APOEε4 on cognitive function, although results are mixed. 2. Interpretation: In a large, community-based sample of cognitively unimpaired, middle-aged adults, OSA was associated with poorer memory and greater dementia risk. These associations were partly explained by vascular risk factors and were not moderated by APOE ε4 status. These findings suggest that the contribution of OSA to cognitive function and dementia risk in midlife may be independent of APOE ε4 status. 3. Future directions: Prospective studies are needed to determine whether OSA accelerates cognitive decline over time and whether APOE ε4 modifies these trajec- tories as individuals age. Randomized controlled trials are also needed to determine whether treating OSA in midlife can reduce vascular burden and ultimately mitigate dementia risk. determine levels of daytime sleepiness; the Insomnia Severity Index (ISI),41 to assess insomnia symptoms and severity; the Pittsburgh Sleep Quality Index (PSQI),42 to assess sleep quality;42 and the Interna- tional Physical Activity Questionnaire (IPAQ),43 to determine physical activity levels.2.3 OSA statusParticipants who self-reported a clinical diagnosis of OSA were assigned to the OSA+ group, while those who did not comprised the OSA− group. Participants in the OSA+ group also indicated whether they were currently receiving treatment for their OSA (treated/untreated).2.4 CognitionParticipants completed the Cogstate Brief Battery, which has been validated for unsupervised, online administration.37,44,45 As described previously,46,47 the battery comprises four subtests: Detection, Iden- tification, One-Back, and One-Card Learning. Briefly, Detection mea- sured psychomotor processing speed through a simple reaction time task; Identification evaluated visual attention via a choice reaction time paradigm; One-Back assessed working memory using a one-back task; and One-Card Learning evaluated visual learning through a continu- ous recognition paradigm within a pattern separation model. Reaction time (milliseconds) was the primary outcome for Detection and Iden- tification, which was log10-transformed and reversed so that higher scores reflected better performance. Accuracy was the primary out- come for One-Back and One-Card Learning, which was arcsine square root transformed, with higher scores reflecting better accuracy. All raw scores were standardized using the sample mean and standard deviation. Composites for attention and memory were computed, respectively, by averaging standardized Detection and Identification scores and standardized One-Back and One-Card Learning scores. These composites have demonstrated sensitivity to mild cognitive impairment (MCI) and AD, with studies suggesting greater sensitivity compared to individual subtests.46–48 Distribution of z-scores for each cognitive composite are shown in Figure S1.2.5 Dementia riskThe CAIDE risk score has been previously validated as an effec- tive tool for predicting late-life cognitive impairment and dementia in midlife.49,50 The score (without APOE ε4) incorporates seven risk factors: age, years of education, sex, BMI (kg/m2), physical activity, his- tory of hypercholesterolemia, and history of hypertension (Table S1). Total scores ranged from 0 to 15, with higher scores indicating greater dementia risk. To isolate dementia risk attributable to vascular risk factors both associated with OSA and readily modifiable, a modified version of the CAIDE risk score was also computed, including only measures of BMI, physical activity, history of hypercholesterolemia, and history of hypertension.51–54 Modified CAIDE risk scores ranged from 0 to 7, with higher scores reflecting greater vascular risk burden.2.6 APOE genotypingGenotek Oragene (OG-500) 2 mL saliva kits were mailed to partic- ipants who completed at least 80% of assessments on the larger HBP online platform.37 Returned saliva samples were forwarded to GenoFIND Services Laboratory (Salt Lake City, USA), where they were processed to extract targeted SNPs, including APOE (rs429358, rs7412), using the TaqMan GTXpress Master Mix (Life Technologies) methodology according to manufacturer instructions.2.7 Statistical analysesAnalyses were performed in R (version 4.4.0). Demographic and health characteristics were compared between the OSA+ and OSA− groups using t-tests for continuous variables and χ2 tests for categorical variables. To examine between-group differences in cognition, dementia risk, and modifiable vascular dementia risk, a series of analyses of covari- ### Page 4 4 of 12 ABDELMESSIH ET AL.Participants enrolled in the HBP by December 2022 (fifth data freeze) (N = 6,912) Complete demographic and sleep health history data (n = 3,145) Final analysis sample (n = 2,795) Excluded (n = 3,767):• Did not provide complete demographic information (i.e., age, sex, education) (568)• Did not provide a sleep health history (3,199) Excluded (n = 350):• Self-reported sleep disorder other than OSA (e.g., insomnia, restless leg syndrome) (350) Cogstate Brief Battery sample (n = 1,830) OSA+ group (n = 127) OSA– group (n = 1,703) OSA+ group (n = 160) OSA– group (n = 2,173) CAIDE risk score sample (n = 2,333)F I G U R E 1 Participant selection process. ance (ANCOVAs) were performed, with OSA group (OSA−, OSA+) as a fixed factor. Covariates included age (years), sex (male/female), education (years), and OSA treatment status (yes/no). For analyses involving the original CAIDE risk score, age, sex, and education (years), were not included as covariates, as these variables are components of the score itself. To determine whether these associations were modified by APOE ε4 carriage, OSA group × APOE ε4 status interaction terms were added to each model. Planned comparisons were conducted to quantify: (1) the individual contribution of OSA in the absence of ε4 (OSA+/ε4− versus OSA−/ε4−), (2) the individual contribution of APOE ε4 in the absence of OSA (OSA−/ε4+ versus OSA−/ε4−), and (3) whether the combined risk of OSA and APOE ε4 (OSA+/ε4+) was associated with poorer cogni- tion and greater dementia risk compared to each individual risk group (OSA+/ε4−, OSA−/ε4+, OSA−/ε4−). Estimated marginal means were calculated for each group for all comparisons, with the magnitude of difference between groups quan- tified using Cohen’s d effect sizes with 95% confidence intervals (CIs). False discovery rate (FDR) correction for multiple comparisons was applied using the Benjamini−Hochberg method.55 Statistical signifi- cance was set to p < 0.05 for all analyses, except for interaction effects, which were considered significant at p < 0.10 due to their typically lower statistical power relative to main effects. 2.7.1 Sensitivity analyses In a first sensitivity analysis, to examine whether associations differed by OSA treatment status, the primary ANCOVA models were repeated with OSA group coded according to treatment status (OSA−, untreated OSA+, and treated OSA+). OSA treatment status was excluded as a covariate in these models. In a second sensitivity analysis, we exam- ined whether cumulative vascular risk, indexed by the modified CAIDE risk score, accounted for the associations between OSA and cognition, given vascular risk factors commonly co-occur with OSA, and are asso- ciated with cognitive function and increased dementia risk.8,56–58 First, the modified CAIDE risk score was included as an additional covariate in the primary ANCOVA models. A formal mediation analysis was also conducted to determine whether the modified CAIDE risk score medi- ated any significant between-group differences in cognition. Indirect ### Page 5 ABDELMESSIH ET AL. 5 of 12TA B L E 1 Demographic characteristics.Mean (SD) or No. (%) Characteristics Total sample (N = 2795) OSA– group(n = 2600) OSA+ (n = 195) p-ValueAge, years 56.44 (7.09) 56.28 (7.12) 58.49 (6.46) <0.001 Female 2123 (76.0%) 2014 (77.5%) 109 (55.9%) <0.001 Education, years 16.11 (3.43) 16.16 (3.41) 15.46 (3.68) 0.01 Caucasian 2296 (82.2%) 2141 (82.4%) 155 (79.5%) 0.34APOE ε4 carrier* 706 (33.5%) 650 (33.2%) 56 (37.8%) 0.50 First- or second-degree family history of dementia 1495 (55.4%) 1390 (55.3%) 105 (57.1%) 0.64 BMI, kg/m2 26.62 (5.52) 26.34 (5.43) 30.31 (5.45) <0.001 Hypertension 636 (22.8%) 545 (21.0%) 91 (46.7%) <0.001 Hypercholesterolemia 827 (29.6%) 739 (28.4%) 88 (45.1%) <0.001 HADS†Anxiety, total score 4.13 (3.36) 4.08 (3.34) 4.73 (3.64) 0.01 Depression, total score 2.54 (3.11) 2.45 (3.03) 3.80 (3.85) <0.001 Physical activity 0.08 Low 263 (10.9%) 238 (10.6%) 25 (15.2%) Moderate 1028 (42.6%) 954 (42.4%) 74 (45.1%) High 1122 (46.5%) 1057 (47.0%) 65 (39.6%) PSQI, total score‡ 5.38 (3.36) 5.45 (3.40) 6.45 (3.82) <0.001 ESS, total score 5.22 (3.69) 5.10 (3.63) 6.87 (4.15) <0.001 ISI, total score¶ 6.26 (5.02) 6.15 (4.95) 7.67 (5.72) <0.001 Abbreviations: APOE, apolipoprotein E; BMI, body mass index; ESS, Epworth Sleepiness Scale; HADS, Hospital Anxiety and Depression Scale; ISI, Insomnia Severity Index; OSA, obstructive sleep apnea; PSQI, Pittsburgh Sleep Quality Index. *Sample size of participants with available APOE genotyping data = 2106.†Score range, 0 to 21 points for each subscale (anxiety, depression), with scores > 8 indicating clinically significant symptoms.‡Score range, 0 to 21, with scores > 5 indicating poor sleep quality. §Score range, 0 to 24 points, with scores > 10 indicating excessive daytime sleepiness. ¶Score range, 0 to28 points, with scores > 10 clinically significant insomnia symptoms. effects were tested using a 95% bias-corrected CI derived from 1000 bootstrap samples.3 RESULTS 3.1 Demographic characteristicsOf the 2795 participants (mean [SD] age = 56.44 [7.09]; 76% female), 195 (7%) reported a diagnosis of OSA (OSA+) and 2600 (93%) did not (OSA−). The demographic and health characteristics of the sample stratified by OSA group are summarized in Table 1. APOE ε4 carriers comprised 33.5% of the sample, and 55.4% had a first- or second- degree family history of dementia. On average, compared to the OSA−group, the OSA+ group was older, more likely to be men, had a higher BMI, fewer years of education, greater depression and anxiety symp- toms, and a higher proportion of vascular risk factors. The OSA+ group also reported greater daytime sleepiness, poorer sleep quality, more severe insomnia symptoms, and more frequently endorsed both stop- ping breathing and snoring or coughing loudly during sleep (28% vs. 6%). Of the OSA+ group, 58.9% reported current treatment for their OSA.3.2 OSA and cognitionAfter adjustments for age, sex, education, and OSA treatment, partic- ipants in the OSA+ group demonstrated poorer performance on the memory composite compared to those in the OSA− group (Table 2). The magnitude of this difference was moderate by convention. No between-group differences were observed in performance on the attention composite.3.3 OSA and dementia riskParticipants in the OSA+ group had a higher original CAIDE risk score compared to those in the OSA− group after adjustment for OSA treatment (Table 2). The magnitude of this difference was large by convention. These results were comparable when using the modi- ### Page 6 6 of 12 ABDELMESSIH ET AL.TA B L E 2 Associations of OSA with cognition and dementia risk.ANCOVA EMM (SE) OSA+ versus OSA– Parameter β (SE) p-Value OSA– OSA+ Cohen’s d (95% CI) p-ValueAttention composite −0.13 (0.11) 0.52 0.02 (0.02) −0.04 (0.08) −0.06 (−0.25, 0.12) 0.49 Memory composite −0.27 (0.10) 0.01 −0.00 (0.02) −0.17 (0.07) −0.22 (−0.04, −0.40) 0.03 CAIDE risk score 1.62 (0.29) <0.001 5.08 (0.19) 7.04 (0.05) 0.84 (0.68, 1.00) <0.001 Modified CAIDE risk score 0.87 (0.21) <0.001 1.50 (0.04) 2.72 (0.14) 0.72 (0.56, 0.89) <0.001 Abbreviations: ANCOVA, analysis of covariance; CAIDE, Cardiovascular Risk Factors, Ageing and Incidence of Dementia; CI, confidence interval; EMM, estimated marginal mean; OSA, obstructive sleep apnea; SE, standard error.Note: Models adjusted for age, sex, education, and OSA treatment where appropriate. p-Values are corrected for multiple comparisons. Sample sizes: Atten- tion composite, N = 1830 (OSA− = 1703, OSA+ = 127); Memory composite, N = 1791 (OSA− = 1666, OSA+ = 125); CAIDE risk scores, N = 2333 (OSA− =2173, OSA+ = 160).F I G U R E 2 The magnitude of difference in attention, memory, dementia risk, and modifiable vascular dementia risk between APOE ε4 non-carriers without OSA and those with APOE ε4 only, OSA only and OSA and APOE ε4. Error bars represent 95% confidence intervals. * Indicatesp < 0.01. ** Indicates p < 0.001. APOE, apolipoprotein E; OSA, obstructive sleep apnea. fied CAIDE risk score, although the magnitude of this difference was moderate by convention.3.4 Modification by APOE ε4 carriageSignificant interactions between OSA and APOE ε4 were observed for the original CAIDE risk score (F2-2326 = 5.99; p = 0.01) and modified CAIDE risk score (F2-2323 = 3.50; p = 0.06). Pairwise compar- isons indicated that, compared to participants with neither risk factor (OSA−/ε4−), those with OSA only (OSA+/ε4−) and those with both OSA and APOE ε4 (OSA+/ε4+) had significantly higher original and modified CAIDE risk scores, while those with APOE ε4 only (OSA−/ε4+) did not (Table 3 and Figure 2). Participants with both OSA and APOEε4 had higher original and modified CAIDE scores than those withAPOE ε4 only (OSA−/ε4+), but not compared to those with OSA only (OSA+/ε4−) (Figure 3). No significant OSA × APOE ε4 interactions were observed for the attention composite (F2-1820 = 1.73; p = 0.24) and memory composite (F2-1781 = 1.30; p = 0.27), and planned comparisons revealed no sig- nificant group differences in attention or memory performance across OSA/APOE ε4 subgroups (Table 3 and Figure 2).3.5 Sensitivity analysesWhen OSA status was further stratified by treatment, poorer mem- ory performance was observed among individuals with untreated OSA ### Page 7 ABDELMESSIH ET AL. 7 of 12F I G U R E 3 Bar plots indicate mean (A) CAIDE risk scores and (B) modified CAIDE risk scores across OSA/APOE ε4 groups. Error bars represent standard error. * Indicates p < 0.01. ** Indicates p < 0.001. APOE, apolipoprotein E; CAIDE, Cardiovascular Risk Factors, Ageing and Incidence of Dementia; OSA, obstructive sleep apnea. compared with those without OSA, while those with treated OSA did not differ significantly from the OSA− group. No significant differences in attention performance were observed for either untreated OSA or treated OSA relative to the OSA− group. Both untreated and treated OSA were associated with higher original and modified CAIDE risk scores compared with those without OSA (Table S2). After additional adjustment for the modified CAIDE risk score, the association between OSA and memory performance was no longer significant (β [SE] = −0.18 [0.11]; p = 0.16), and the effect size was reduced to small (d = 0.14; 95% confidence interval [CI], −0.06 to 0.34). The association between OSA and attention performance remained nonsignificant (β [SE] = −0.07 [0.12]; p = 0.91; d = 0.01; 95% CI, −0.18–0.21). In the mediation model for memory, partici- pants with OSA had significantly higher modified CAIDE risk scores compared to those without OSA (path a: β = 0.83; 95% CI, 0.36 to 1.30; p < 0.001), but modified CAIDE risk scores were not sig- nificantly associated with memory performance (path b: β = −0.02; 95% CI, −0.05 to 0.01; p = 0.17). The indirect effect showed mod- ified CAIDE risk scores did not mediate the association between OSA and memory performance (β = −0.01; 95% CI, −0.04 to 0.01;p = 0.21).4 DISCUSSIONThe hypothesis that participants with OSA would have poorer cog- nition was partially supported. Participants with OSA demonstrated poorer memory, with the magnitude of this difference moderate by convention. No between-group differences in attention were observed. Previous studies have reported poorer memory perfor- mance in individuals with OSA compared to those without OSA59–61; however, most studies to date have not accounted for the role of vas- cular risk in these associations.22,62 Given that vascular risk factors such as obesity, hypertension, and hypercholesterolemia frequently co-occur with OSA,56 and are themselves associated with memory impairment,58 it was important to determine whether OSA was inde- pendently associated with cognition beyond its vascular burden. In the current study, while the association between OSA and memory was attenuated after adjusting for vascular risk, a formal mediation anal- ysis did not support a direct or indirect effect of vascular risk factors on memory. These results suggest that, although vascular risk might explain some of the association between OSA and poor memory in midlife, the association may not be fully mediated by vascular burden and may instead reflect underlying vulnerability that modifies suscep- tibility to OSA-related cognitive effects. Notably, sensitivity analyses further indicated that poorer memory performance was evident pri- marily among individuals with untreated OSA, whereas individuals receiving treatment did not differ significantly from those without OSA. Although these findings should be interpreted cautiously given the cross-sectional design and reliance on self-reported binary treat- ment status, they are consistent with the possibility that effective OSA management may attenuate the impact of OSA on memory, even in the presence of elevated vascular risk. Thus, it is likely that nonvascular mechanisms, such as intermittent hypoxemia-related medial temporal lobe atrophy,63 also partly underlie the association between OSA and poor memory. ### Page 8 8 of 12 ABDELMESSIH ET AL.TA B L E 3 Differences in cognition and dementia risk by OSA and APOE ε4 status.Parameter OSA–/ε4– versus OSA–/ε4+ OSA–/ε4– versus OSA+/ε4– OSA–/ε4– versus OSA+/ε4+ OSA+/ε4+ versus OSA–/ε4+ OSA+/ε4+ versus OSA+/ε4–β (SE) Cohen’s d (95% CI) p-Value β (SE) Cohen’s d (95% CI) p-Value β (SE) Cohen’s d (95% CI) p-Value β (SE) Cohen’s d (95% CI) p-Value β (SE) Cohen’s d (95% CI) p-ValueAttention composite−0.016 (0.05) −0.02 (-0.13, 0.10) 0.99 0.03 (0.11) −0.04 (−0.28, 0.21) 0.99 0.06 (0.14) 0.07 (−0.26, 0.38) 0.99 −0.07 (0.14) 0.08 (−0.24, 0.41) 0.99 0.09 (0.17) 0.10 (−0.29, 0.50) 0.99 Memory composite−0.08 (0.04) −0.11 (−0.22, 0.00) 0.99 0.11 (0.1) −0.14 (−0.39, 0.10) 0.99 −0.21 (0.13) −0.27 (−0.59, 0.05) 0.99 0.13 (0.13) −0.16 (−0.49, 0.16) 0.99 −0.10 (0.16) −0.13 (−0.52, 0.27) 0.99 CAIDE risk score−0.14 (0.12) −0.06 (−0.16, 0.04) 0.99 −2.52 (0.26) 1.09 (0.86, 1.31) <0.001 1.50 (0.33) 0.64 (0.36, 0.91) <0.001 −1.61 (0.33) 0.70 (0.41, 0.98) <0.001 −1.05 (0.41) −0.45 (−0.80, −0.10) 0.29 Modified CAIDE risk score−0.08 (0.09) −0.05 (−0.15, 0.05) 0.99 −1.55 (0.19) 0.92 (0.69, 1.14) <0.001 0.94 (0.24) 0.56 (0.28, 0.83) <0.01 −1.02 (0.25) 0.60 (0.32, 0.89) <0.01 −0.61 (0.30) −0.36 (−0.71, −0.01) 0.99 Abbreviations: APOE, apolipoprotein E; CAIDE, Cardiovascular Risk Factors, Ageing and Incidence of Dementia; CI, confidence interval; OSA, obstructive sleep apnea; SE, standard error.Note: Models adjusted for age, sex, education, and OSA treatment where appropriate. p-Values are corrected for multiple comparisons. The hypothesis that OSA would be associated with greater demen- tia risk was also supported. Compared to those without OSA, individu- als with OSA showed significantly higher dementia risk, as measured by the CAIDE risk score, with the magnitude of this difference large by convention. This finding is consistent with prospective studies that show individuals with OSA are more likely to develop cognitive impairment,2,3 and extend prior work by demonstrating that this asso- ciation is evident even in a relatively young, cognitively unimpaired sample. Notably, the association remained significant when the CAIDE risk score was restricted to modifiable vascular factors (i.e., hyperten- sion, hypercholesterolemia, physical inactivity, obesity). This observa- tion is consistent with prior research by our group64,65 and others66 that report a link between OSA and the cumulative burden of vas- cular risk factors, and supports broader literature implicating OSA in the development of cardiovascular and cerebrovascular risk.7,56,67–69 The mechanisms underlying this relationship are well-documented; intermittent hypoxia and sleep fragmentation may activate the sym- pathetic activity and renin-angiotensin-aldosterone system, leading to endothelial dysfunction, increased arterial stiffness, and chronic inflammation, all of which likely promote the development of vascu- lar risk factors, such as hypertension.69 Importantly, elevated vascular dementia risk was observed in both treated and untreated OSA groups compared with those without OSA, suggesting that an OSA diagnosis in midlife may reflect an accumulated burden of vascular risk factors that may not be fully attenuated cross-sectionally by treatment. Taken together, these findings add further support to the hypothesis that vascular mechanisms contribute to the association between OSA and late-life cognitive impairment and dementia.APOE ε4 carriage did not moderate the relationship between OSA and cognition in the current sample. Specifically, individuals with both OSA and APOE ε4 did not have poorer memory than the other sub- groups, and the magnitude of group differences was small to moderate by convention (d’s = 0.11–0.27). These findings contrast with previ- ous cross-sectional studies that have reported an interaction between OSA and APOE ε4 on cognition in middle-aged adults across both community-based and clinical samples.14,31,33 There are several poten- tial reasons for this discrepancy. First, APOE ε4-related cognitive deficits are more consistently observed later in life.70 In middle-aged adults, ε4-related cognitive deficits have primarily been observed inAPOE ε4 homozygotes rather than heterozygotes.71 Given the rela- tively young age of the current sample and the small number of APOE ε4 homozygotes (i.e., n = 74), it is likely that the current study was under- powered to adequately test OSA × APOE ε4 interactions on cognition at cross-section. Second, the interactive effect of OSA and APOE ε4 on cognition has been primarily observed in cases of severe OSA.31–35 As HBP participants were required to self-report their OSA status, infor- mation on OSA severity (e.g., Apnea-Hypopnea Index, which can only be determined through polysomnography) was unavailable. The cur- rent sample also included participants who were already undergoing OSA treatment (58.9%), likely resulting in a sample with milder disease severity overall compared to previous studies. While individuals with both OSA and APOE ε4 exhibited elevated dementia risk, those with OSA alone showed the greatest dementia ### Page 9 ABDELMESSIH ET AL. 9 of 12risk. Conversely, APOE ε4 carriers without OSA did not differ from those with neither risk factor. This pattern is consistent with the notion that, in midlife, the vascular and cerebrovascular effects of APOE ε4 have not yet manifested fully, and instead accumulate gradually and become more apparent in later life.29,72,73 In cognitively unimpaired older adults, APOE ε4 and vascular risk have both been independently associated with greater hippocampal atrophy and episodic memory decline,54 while other work has shown APOE ε4 moderates the asso- ciation between vascular risk factors and tau pathology.74 When considered together, vascular and genetic risk may operate indepen- dently earlier in the disease course but increasingly converge as age and pathology accumulate. Thus, while the present cross-sectional findings suggest an early and independent contribution of OSA to dementia risk and vascular burden, it remains possible that the addi- tive or synergistic effects of OSA and APOE ε4 emerge over time. Future studies will need to determine whether interactions between OSA andAPOE ε4 on cognition and dementia risk are more evident in individ- uals who are older, have greater pathological burden (e.g., amyloid-β, tau, cerebrovascular disease), or have more severe OSA. It will also be important to clarify whether elevated dementia risk in midlife OSA is associated with subsequent cognitive decline, and whether APOE ε4 moderates these trajectories.4.1 Strengths and limitationsAn important strength of this study was the large, well-characterized, community-based sample, over half of whom reported a family his- tory of dementia. Prior studies have largely relied on sleep clinic samples,75 which may be biased toward individuals with more severe symptomatology and greater comorbidity burden and overall dementia risk. Therefore, our community-based cohort may offer better insight into a broader spectrum of OSA severity and its associated risk pro- file in community-dwelling adults. This enhances the generalizability of the current findings to the broader population, which is particularly relevant for population-level dementia prevention strategies and sec- ondary prevention trials (e.g., BetterBrains,76 Maintain Your Brain77). Further, as late-life OSA may also represent symptoms of incipient dementia,78 the examination of these associations in midlife reduces the likelihood that the results of the current study reflect reverse causation. There are also several limitations that should be considered. First, OSA status was determined by self-reported diagnosis, which may have resulted in misclassification, especially given OSA is frequently underdiagnosed.79 Although the proportion of OSA cases in this sample (i.e., 7%) is comparable to previous population estimates for middle-aged adults (e.g., 9%),80 some individuals with undiag- nosed OSA were likely included in the OSA− group, which would be expected to bias associations toward the null. The reliance on self-report also precluded examination of objective markers of OSA severity (e.g., Apnea-Hypopnoea Index, hypoxemic burden). Nonethe- less, this approach may have preferentially captured individuals with more symptomatic OSA sufficient to prompt clinical diagnoses.81 Sec- ond, detailed information regarding OSA treatment type, duration, and adherence was unavailable, limiting further inferences regard- ing the potential modifying effects of treatment. Third, while the Cogstate Brief Battery is validated and sensitive to early AD-related cognitive changes,46–48 it is not as comprehensive as a full neu- ropsychological assessment. This may have also limited the ability to detect subtle OSA− or APOE ε4-related impairments in other cogni- tive domains (e.g., verbal memory, executive function). Finally, owing to the cross-sectional design, conclusions regarding the temporal rela- tionship between OSA and cognitive decline, or about whether these associations differ prospectively by APOE ε4 status, could not be drawn.5 CONCLUSIONThese limitations notwithstanding, the results of this study suggest that, in middle-aged adults, OSA is associated with subtle yet mea- surable memory impairment and greater dementia risk, irrespective of APOE ε4 carriage. These findings also imply a dissociation between OSA-related cognitive vulnerability and dementia risk in midlife, whereby subtle memory impairment is most evident in untreated OSA, while elevated dementia risk reflects cumulative vascular burden irre- spective of treatment status. As such, routine screening for OSA in middle-aged adults may help identify individuals at risk of dementia and provide an opportunity for intervention before clinically significant cognitive impairment emerges. Future randomized controlled trials are needed to investigate whether targeting OSA together with associ- ated vascular comorbidities could potentially slow or prevent cognitive decline or reduce dementia risk in later life.AUTHOR CONTRIBUTIONS Gabriel T. Abdelmessih: Conceptualization; formal analysis; methodol- ogy; writing—original draft; writing—review and editing. Lisa Bransby: Writing—review and editing. Hannah Cummins: Writing—review and editing. Melinda L. Jackson: Conceptualization; supervision; writing— review and editing. Yen Ying Lim: Conceptualization; funding acquisi- tion; supervision; writing—review and editing.ACKNOWLEDGMENTSWe thank the HBP participants for their commitment to helping us bet- ter understand the risk and protective factors of dementia. The HBP (healthybrainproject.org.au) was funded by the National Health and Medical Research Council (NHMRC; GNT1158384, GNT1147465, GNT1111603, GNT1105576, GNT1104273, GNT1171816), the Alzheimer’s Association (AARG-17-591424, AARG-18-591358, AARG-19-643133), the Dementia Australia Research Foundation, the Bethlehem Griffiths Research Foundation, the Yulgilbar Alzheimer’s Research Program, the National Heart Foundation of Australia (102052), and the Charleston Conference for Alzheimer’s Disease. The authors thank their study partners (PearlArc, SRC Innovations, Cogstate Ltd., and Cambridge Cognition) for their support. G.T. Abdelmessih is supported by an Australian Government Research ### Page 10 10 of 12 ABDELMESSIH ET AL.Training Program (RTP) Scholarship. Y.Y Lim is supported by an NHMRC Emerging Leadership 2 grant (GNT2009550). Open access publishing facilitated by Monash University, as part of the Wiley - Monash University agreement via the Council of Australasian University LibrariansCONFLICT OF INTEREST STATEMENTAll authors have no relevant disclosures to report. Author disclosures are available in the Supporting Information.CONSENT STATEMENTAll participants provided informed consent via the HBP online platform prior to completing the online assessments.ORCIDGabriel T. Abdelmessih https://orcid.org/0000-0001-5334-8535REFERENCES1. Jordan AS, McSharry DG, Malhotra A. Adult obstructive sleep apnoea. Lancet. 2014;383(9918):736-747. doi:10.1016/S0140- 6736(13)60734-5 2. Bubu OM, Brannick M, Mortimer J, et al. Sleep, cognitive impairment, and Alzheimer’s disease: a systematic review and meta-analysis. Sleep. 2017;40(1): doi:10.1093/sleep/zsw032 3. Leng Y, McEvoy CT, Allen IE, Yaffe K. Association of sleep- disordered breathing with cognitive function and risk of cognitive impairment: a systematic review and meta-analysis. JAMA Neu- rol. 2017;74(10):1237-1245. doi:10.1001/jamaneurol.2017. 2180 4. Liguori C, Maestri M, Spanetta M, et al. Sleep-disordered breathing and the risk of Alzheimer’s disease. Sleep Med Rev. 2021;55:101375. doi:10.1016/j.smrv.2020.101375 5. Gosselin N, Baril AA, Osorio RS, Kaminska M, Carrier J. Obstruc- tive sleep apnea and the risk of cognitive decline in older adults.Am J Respir Crit Care Med. 2019;199(2):142-148. doi:10.1164/rccm. 201801-0204PP 6. Andrade AG, Bubu OM, Varga AW, Osorio RS. The relationship between obstructive sleep apnea and Alzheimer’s disease. J Alzheimers Dis. 2018;64(S1):S255-S270. doi:10.3233/JAD-179936 7. Sánchez-de-la-Torre M, Campos-Rodriguez F, Barbé F. Obstruc- tive sleep apnoea and cardiovascular disease. Lancet Respir Med. 2013;1(1):61-72. doi:10.1016/S2213-2600(12)70051-6 8. Xu H, Guan J, Yi H, Zou J et al. Elevated low-density lipopro- tein cholesterol is independently associated with obstructive sleep apnea: evidence from a large-scale cross-sectional study. Sleep Breath. 2016;20(2):627-634. doi:10.1007/s11325-015-1262-3 9. Fattal D, Hester S, Wendt L. Body weight and obstructive sleep apnea: a mathematical relationship between body mass index and apnea- hypopnea index in veterans. J Clin Sleep Med. 2022;18(12):2723-2729. doi:10.5664/jcsm.10190 10. Simpson L, McArdle N, Eastwood PR, et al. Physical inactivity is asso- ciated with moderate-severe obstructive sleep apnea. J Clin Sleep Med. 2015;11(10):1091-1099. doi:10.5664/jcsm.5078 11. Ju G, Yoon IY, Lee SD, Kim TH, Choe JY, Kim KW. Effects of sleep apnea syndrome on delayed memory and executive function in elderly adults. J Am Geriatr Soc. 2012;60(6):1099-1103. doi:10.1111/j.1532- 5415.2012.03961.x 12. Naëgelé B, Launois SH, Mazza S, Feuerstein C, Pépin JL, Lévy P. Which memory processes are affected in patients with obstructive sleep apnea? An evalation of 3 types of memory. Sleep. 2006;29(4):533-544. doi:10.1093/sleep/29.4.533 13. Saunamäki T, Himanen SL, Polo O, Jehkonen M. Executive dysfunc- tion in patients with obstructive sleep apnea syndrome. Eur Neurol. 2009;62(4):237-242. doi:10.1159/000232156 14. Lui KK, Dave A, Sprecher KE, et al. Older adults at greater risk for Alzheimer’s disease show stronger associations between sleep apnea severity in REM sleep and verbal memory. Alzheimers Res Ther. 2024;16(1):102. doi:10.1186/s13195-024-01446-3 15. Pase MP, Harrison S, Misialek JR, et al. Sleep architecture, obstruc- tive sleep apnea, and cognitive function in adults. JAMA Netw Open. 2023;6(7):e2325152. doi:10.1001/jamanetworkopen.2023.25152 16. Yaffe K, Laffan AM, Harrison SL, et al. Sleep-disordered breathing, hypoxia, and risk of mild cognitive impairment and dementia in older women. JAMA. 2011;306(6):613-619. doi:10.1001/jama.2011.1115 17. Haba-Rubio Jé, Marti-Soler H, Tobback N, et al. Sleep characteris- tics and cognitive impairment in the general population. Neurology. 2017;88(5):463-469. doi:10.1212/WNL.0000000000003557 18. Shieu MM, Dunietz GL, Paulson HL, Chervin RD, Braley TJ. The asso- ciation between obstructive sleep apnea risk and cognitive disorders: a population-based study. J Clin Sleep Med. 2022;18(4):1177-1185. doi:10.5664/jcsm.9832 19. Thompson C, Legault J, Moullec G, et al. Association between risk of obstructive sleep apnea, inflammation and cognition after 45 years old in the Canadian longitudinal study on aging. Sleep Med. 2022;91:21-30. doi:10.1016/j.sleep.2022.02.006 20. Sforza E, Roche F, Thomas-Anterion C, et al. Cognitive function and sleep-related breathing disorders in a healthy elderly population: the SYNAPSE Study. Sleep. 2010;33(4):515-521. doi:10.1093/sleep/33.4. 515 21. Lutsey PL, Bengtson LGS, Punjabi NM, et al. Obstructive sleep apnea and 15-year cognitive decline: the Atherosclerosis Risk in Commu- nities (ARIC) Study. Sleep. 2016;39(2):309-316. doi:10.5665/sleep. 5434 22. Boland LL, Shahar E, Iber C, Knopman DS, Kuo TF, Nieto FJ Measures of cognitive function in persons with varying degrees of sleep-disordered breathing: the Sleep Heart Health Study. J Sleep Res. 2002;11(3):265- 272. doi:10.1046/j.1365-2869.2002.00308.x 23. Dlugaj M, Weinreich G, Weimar C, et al. Sleep-disordered breathing, sleep quality, and mild cognitive impairment in the general population.J Alzheimers Dis. 2014;41(2):479-497. doi:10.3233/JAD-132132 24. Scheltens P, De Strooper B, Kivipelto M, et al. Alzheimer’s disease.Lancet. 2021;397(10284):1577-1590. doi:10.1016/S0140-6736(20) 32205-4 25. Lim YY, Mormino EC, Weiner M, et al. APOE genotype and early beta- amyloid accumulation in older adults without dementia. Neurology. 2017;89(10):1028-1034. doi:10.1212/WNL.0000000000004336 26. Burnham SC, Laws SM, Budgeon CA, et al. Impact of APOE-ε4 carriage on the onset and rates of neocortical Aβ-amyloid deposition. Neurobiol Aging. 2020;95:46-55. doi:10.1016/j.neurobiolaging.2020.06.001 27. Villemagne VL, Burnham S, Bourgeat P, Brown B et al. Amyloid beta deposition, neurodegeneration, and cognitive decline in spo- radic Alzheimer’s disease: a prospective cohort study. Lancet Neurol. 2013;12(4):357-367. doi:10.1016/S1474-4422(13)70044-9 28. Sutphen CL, Jasielec MS, Shah AR, et al. Longitudinal cerebrospinal fluid biomarker changes in preclinical Alzheimer disease during middle age. JAMA Neurol. 2015;72(9):1029-1042. doi:10.1001/jamaneurol. 2015.1285 29. Zlokovic BV. Cerebrovascular effects of apolipoprotein E: implica- tions for Alzheimer disease. JAMA Neurol. 2013;70(4):440-444. doi:10. 1001/jamaneurol.2013.2152 30. Montagne A, Nation DA, Sagare AP, et al. APOE4 leads to blood- brain barrier dysfunction predicting cognitive decline. Nature. 2020;581(7806):71-76. doi:10.1038/s41586-020-2247-3 31. Cosentino FII, Bosco P, Drago V, et al. The APOE ε4 allele increases the risk of impaired spatial working memory in obstructive sleep apnea.Sleep Med. 2008;9(8):831-839. doi:10.1016/j.sleep.2007.10.015 ### Page 11 ABDELMESSIH ET AL. 11 of 1232. Marchi NA, Berger M, Solelhac G, et al. Obstructive sleep apnea and cognitive functioning in the older general population: the moderating effect of age, sex, ApoE4, and obesity. J Sleep Res. 2024;33(1):e13938. doi:10.1111/jsr.13938 33. Nikodemova M, Finn L, Mignot E, Salzieder N, Peppard PE. Association of sleep-disordered breathing and cognitive deficit in APOE ε4 carriers.Sleep. 2013;36(6):873-880. doi:10.5665/sleep.2714 34. Spira AP, Blackwell T, Stone KL, et al. Sleep-disordered breathing and cognition in older women. J Am Geriatr Soc. 2008;56(1):45-50. doi:10. 1111/j.1532-5415.2007.01506.x 35. Johnson DA, Lane J, Wang R, et al. Greater cognitive deficits with sleep-disordered breathing among individuals with genetic suscepti- bility to Alzheimer disease. Ann Am Thorac Soc. 2017;14(11):1697- 1705. doi:10.1513/AnnalsATS.201701-052OC 36. Livingston G, Huntley J, Liu KY, et al. Dementia prevention, interven- tion, and care: 2024 report of the Lancet standing commission. Lancet. 2024;404(10452):572-628. doi:10.1016/S0140-6736(24)01296-0 37. Lim YY, Yassi N, Bransby L, Properzi M, Buckley R. The Healthy Brain Project: an online platform for the recruitment, assessment, and moni- toring of middle-aged adults at risk of developing Alzheimer’s disease.J Alzheimers Dis. 2019;68(3):1211-1228. doi:10.3233/JAD-181139 38. Field N, Cohen T, Struelens MJ, et al. Strengthening the reporting of molecular epidemiology for infectious diseases (STROME-ID): an extension of the STROBE statement. Lancet Infect Dis. 2014;14(4):341- 352. doi:10.1016/S1473-3099(13)70324-4 39. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361-370. doi:10.1111/j.1600-0447.1983. tb09716.x 40. Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14(6):540-545. doi:10.1093/ sleep/14.6.540 41. Morin CM, Belleville Gè, Bélanger L, Ivers H. The insomnia severity index: psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep. 2011;34(5):601-608. doi:10.1093/sleep/ 34.5.601 42. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pitts- burgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193-213. doi:10.1016/0165- 1781(89)90047-4 43. Hagströmer M, Oja P, Sjöström M. The International Physical Activ- ity Questionnaire (IPAQ): a study of concurrent and construct validity.Public Health Nutr. 2006;9(6):755-762. doi:10.1079/PHN2005898 44. Maruff P, Bransby L, Yassi N, Buckley RF, Lim YY. Remote assess- ment of cognition with the unsupervised version of the Cogstate Brief Battery: association of composite endpoints with Alzheimer’s disease biomarkers. Alzheimers Dement. 2023;19(S18):e079309. doi:10.1002/ alz.079309 45. Perin S, Buckley RF, Pase MP, et al. Unsupervised assessment of cognition in the Healthy Brain Project: implications for web-based reg- istries of individuals at risk for Alzheimer’s disease. Alzheimers Dement. 2020;6(1):e12043. doi:10.1002/trc2.12043 46. Lim YY, Ellis KA, Harrington K, et al. Use of the CogState Brief Battery in the assessment of Alzheimer’s disease related cognitive impairment in the Australian Imaging, Biomarkers and Lifestyle (AIBL) study. J Clin Exp Neuropsychol. 2012;34(4):345-358. doi:10.1080/13803395.2011. 643227 47. White JP, Schembri A, Prenn-Gologranc C, et al. Sensitivity of indi- vidual and composite test scores from the Cogstate Brief Battery to mild cognitive impairment and dementia due to Alzheimer’s disease. J Alzheimers Dis. 2023;96(4):1781-1799. doi:10.3233/JAD-230352 48. Maruff P, Lim YY, Darby D, et al. Clinical utility of the Cogstate Brief Battery in identifying cognitive impairment in mild cognitive impair- ment and Alzheimer’s disease. BMC Psychol. 2013;1(1):30. doi:10. 1186/2050-7283-1-30 49. Kivipelto M, Ngandu T, Laatikainen T, Winblad B, Soininen H, Tuomilehto J. Risk score for the prediction of dementia risk in 20 years among middle-aged people: a longitudinal, population- based study. Lancet Neurol. 2006;5(9):735-741. doi:10.1016/S1474- 4422(06)70537-3 50. O’Brien JT, Firbank MJ, Ritchie K, et al. Association between midlife dementia risk factors and longitudinal brain atrophy: the PREVENT- Dementia study. J Neurol Neurosurg Psychiatry. 2020;91(2):158-161. doi:10.1136/jnnp-2019-321652 51. Cavuoto MG, Davies L, Rowsthorn E, et al. Cross-sectional associa- tions between neighborhood characteristics, cognition and dementia risk factor burden in middle-aged and older Australians. Prev Med Rep. 2024;41:102696. doi:10.1016/j.pmedr.2024.102696 52. Franks KH, Bransby L, Cribb L, et al. Associations of perceived stress and psychological resilience with cognition and a modifiable demen- tia risk score in middle-aged adults. J Gerontol B Psychol Sci Soc Sci. 2023;78(12):1992-2000. doi:10.1093/geronb/gbad131 53. Pase MP, Rowsthorn E, Cavuoto MG, et al. Association of neighborhood-level socioeconomic measures with cognition and dementia risk in Australian adults. JAMA Netw Open. 2022;5(3):e224071. doi:10.1001/jamanetworkopen.2022.4071 54. Rosenich E, Bransby L, Yassi N, et al. Differential effects of APOEand modifiable risk factors on hippocampal volume loss and memory decline in Aβ− and Aβ+ older adults. Neurology. 2022;98(17):e1704- e1715. doi:10.1212/WNL.0000000000200118 55. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a prac- tical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol. 1995;57(1):289-300. doi:10.1111/j.2517-6161.1995. tb02031.x 56. Shiina K. Obstructive sleep apnea-related hypertension: a review of the literature and clinical management strategy. Hypertens Res. 2024;47(11):3085-3098. doi:10.1038/s41440-024-01852-y 57. Richards KC, Gooneratne N, Dicicco B, et al. CPAP adherence may slow 1-year cognitive decline in older adults with mild cognitive impair- ment and apnea. J Am Geriatr Soc. 2019;67(3):558-564. doi:10.1111/ jgs.15758 58. Ungvari Z, Toth P, Tarantini S, et al. Hypertension-induced cognitive impairment: from pathophysiology to public health. Nat Rev Nephrol. 2021;17(10):639-654. doi:10.1038/s41581-021-00430-6 59. Stranks EK, Crowe SF. The cognitive effects of obstructive sleep apnea: an updated meta-analysis. Arch Clin Neuropsychol. 2016;31(2):186- 193. doi:10.1093/arclin/acv087 60. Bubu OM, Andrade AG, Umasabor-Bubu OQ, et al. Obstructive sleep apnea, cognition and Alzheimer’s disease: a systematic review inte- grating three decades of multidisciplinary research. Sleep Med Rev. 2020;50:101250. doi:10.1016/j.smrv.2019.101250 61. Wallace A, Bucks RS. Memory and obstructive sleep apnea: a meta-analysis. Sleep. 2013;36(2):203-220. doi:10.5665/sleep. 2374 62. Bawden FC, Oliveira CA, Caramelli P. Impact of obstructive sleep apnea on cognitive performance. Arq Neuropsiquiatr. 2011;69:585- 589. doi:10.1590/S0004-282x2011000500003 63. Canessa N, Castronovo V, Cappa SF, et al. Obstructive sleep apnea: brain structural changes and neurocognitive function before and after treatment. Am J Respir Crit Care Med. 2011;183(10):1419-1426. doi:10. 1164/rccm.201005-0693OC 64. Jackson ML, Cavuoto M, Schembri R, et al. Severe obstructive sleep apnea is associated with higher brain amyloid burden: a preliminary PET imaging study. J Alzheimers Dis. 2020;78(2):611-617. doi:10.3233/ JAD-200571 65. Cavuoto MG, Robinson SR, O’Donoghue FJ, et al. Associations between amyloid burden, hypoxemia, sleep architecture, and cogni- tion in obstructive sleep apnea. J Alzheimers Dis. 2023;96(1):149-159. doi:10.3233/JAD-221049 ### Page 12 12 of 12 ABDELMESSIH ET AL.66. Elias A, Cummins T, Tyrrell R, et al. Risk of Alzheimer’s disease in obstructive sleep apnea syndrome: amyloid-β and tau imaging. J Alzheimers Dis. 2018;66(2):733-741. doi:10.3233/JAD-180640 67. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med. 2000;342(19):1378-1384. doi:10.1056/NEJM200005113421901 68. Qian Y, Yi H, Zou J, et al. Independent association between sleep frag- mentation and dyslipidemia in patients with obstructive sleep apnea.Sci Rep. 2016;6(1):26089. doi:10.1038/srep26089 69. Yeghiazarians Y, Jneid H, Tietjens JR, et al. Obstructive sleep apnea and cardiovascular disease: a scientific statement from the Ameri- can Heart Association. Circulation. 2021;144(3):e56-e67. doi:10.1161/ CIR.0000000000000988 70. O’Donoghue MC, Murphy SE, Zamboni G, Nobre AC, Mackay CE. APOE genotype and cognition in healthy individuals at risk of Alzheimer’s disease: a review. Cortex. 2018;104:103-123. doi:10. 1016/j.cortex.2018.03.025 71. Lim YY, Pase MP, Buckley RF, et al. Visual memory deficits in middle- aged APOE ɛ4 homozygotes detected using unsupervised cognitive assessments. J Alzheimers Dis. 2021;79(4):1563-1573. doi:10.3233/ JAD-201281 72. Stoumpos S, Hamodrakas SJ, Anthopoulos PG, Bagos PG. The asso- ciation between apolipoprotein E gene polymorphisms and essential hypertension: a meta-analysis of 45 studies including 13,940 cases and 16,364 controls. J Hum Hypertens. 2013;27(4):245-255. doi:10.1038/ jhh.2012.37 73. Dunk MM, Li J, Liu S, et al. Associations of dietary cholesterol and fat, blood lipids, and risk for dementia in older women vary by APOEgenotype. Alzheimers Dement. 2023;19(12):5742-5754. doi:10.1002/ alz.13358 74. Tsiknia AA, Reas E, Bangen KJ, et al. Sex and APOE ɛ4 modify the effect of cardiovascular risk on tau in cognitively normal older adults. Brain Commun. 2022;4(1):fcac035. doi:10.1093/braincomms/fcac035 75. Cross N, Lampit A, Pye J, Grunstein RR, Marshall N, Naismith SL. Is obstructive sleep apnoea related to neuropsychological function in healthy older adults? A systematic review and meta-analysis. Neu- ropsychol Rev. 2017;27(4):389-402. doi:10.1007/s11065-017-9344- 6 76. Rosenich E, Ayton D, Perin S, et al. An online, person-centered, risk fac- tor management program to prevent cognitive decline: protocol for a prospective behavior-modification blinded endpoint randomized con- trolled trials. J Alzheimers Dis. 2021;83(4):1603-1622. doi:10.3233/ JAD-210589 77. Brodaty H, Chau T, Heffernan M, et al. An online multidomain lifestyle intervention to prevent cognitive decline in at-risk older adults: a ran- domized controlled trial. Nat Med. 2025;31(2):565-573. doi:10.1038/ s41591-024-03351-6 78. Gottlieb E, Grima N, Howard M, et al. Unraveling the contributions of sleep dysfunction to Alzheimer’s disease. Genetics, Neurology, Behav- ior, and Diet in Dementia. 2020;2:539-552. doi:10.1016/B978-0-12- 815868-5.00034-7 79. Chen X, Wang R, Zee P, et al. Racial/ethnic differences in sleep dis- turbances: the Multi-Ethnic Study of Atherosclerosis (MESA). Sleep. 2015;38(6):877-888. doi:10.5665/sleep.4732 80. Benjafield A, Ayas N, Eastwood P, et al. Prevalence of OSA globally, in Asia and Oceania: estimates using currently available data. J Sleep Res. 2019;28(S1):e49_12912. doi:10.1111/jsr.49_12912 81. Bubu OM, Pirraglia E, Andrade AG, et al. Obstructive sleep apnea and longitudinal Alzheimer’s disease biomarker changes. Sleep. 2019;42(6):zsz048. doi:10.1093/sleep/zsz048SUPPORTING INFORMATIONAdditional supporting information can be found online in the Support- ing Information section at the end of this article.How to cite this article: Abdelmessih GT, Bransby L, Cummins H, Jackson ML, Lim YY. Associations of self-reported obstructive sleep apnea with cognition and dementia risk in cognitively unimpaired middle-aged adults. Alzheimer’s Dement. 2026;22:e71553. https://doi.org/10.1002/alz.71553