Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Oct 7.
Published before final editing as: Gerontologist. 2026 Aug 18:gnag200. doi: 10.1093/geront/gnag200

Midlife physical activity and late-life subjective cognitive complaints in women

Yixiao Song 1,2, Fen Wu 1, Anthony Wong 1, Tess V Clendenen 1, Karen L Koenig 1, Yian Gu 3,4, Anne Zeleniuch-Jacquotte 1, Yu Chen 1,5
PMCID: PMC13640539  NIHMSID: NIHMS2210952  PMID: 42610694

Abstract

Background and Objectives:

Alzheimer’s disease affects mostly women over 65, who make up nearly two-thirds of cases. Subjective cognitive complaints (SCC) can precede Alzheimer’s disease by years. While physical activity supports healthy aging, the long-term impact of midlife physical activity (PA) on late-life SCCs in women is limited.

Research Design and Methods:

Using data from 4,397 participants (mean age=46.2 years at baseline; 78.7 years at SCC assessment) in the New York University Women’s Health Study (NYUWHS), a prospective cohort of 14,274 women enrolled in 1985-1991, we examined the association between midlife PA and SCC reported nearly 30 years later. PA was assessed at baseline using self-reported weekly hours of mild, moderate, and vigorous activity, converted to metabolic equivalent hours per week (MET-hours/week). Odds ratios and 95% CIs were estimated using unconditional logistic regression, adjusting for demographics and midlife and late-life health conditions. Multiple imputation and inverse probability weighting addressed missing data and potential selection bias.

Results:

Women in the highest tertile of total midlife PA had 20% lower odds of reporting ≥2 SCCs compared with those in the lowest tertile (OR=0.80; 95% CI=0.68-0.95). The inverse association was stronger among never smokers (OR=0.68; p for interaction=0.029). The inverse association was consistent for moderate and vigorous, but not mild PA. Findings were consistent across sensitivity analyses using alternative SCC definitions, exclusion criteria and inverse probably weights.

Discussion and Implications:

Higher level of midlife PA was associated with fewer late-life SCCs, suggesting that midlife PA may be an important behavioral correlate of later-life cognitive health in women.

Keywords: Exercise, Alzheimer's disease, Cognitive Dysfunction, Cognition, Diagnostic Self Evaluation, Cohort Studies

Introduction

Alzheimer's disease (AD) is the leading cause of dementia, accounting for over half of all cases and mainly affecting older women aged 65 and above (Prince et al., 2016). Almost two thirds of AD cases are women, highlighting the importance to identify effective modifiable lifestyle-based strategies and protective factors for AD and dementia, especially in women.

Subjective cognitive complaints (SCCs) are self-reported memory and cognitive concerns that may occur before detectable impairment on objective cognitive testing (Jessen et al., 2014). Conceptual frameworks suggest that SCCs may represent an early symptomatic stage along the AD continuum, before clinically apparent mild cognitive impairment (MCI) or dementia (Jessen et al., 2014). Growing evidence indicates that SCCs are associated with an increased risk of subsequent MCI, AD, and dementia, and may also correlate with AD-related pathological changes, such as amyloid-β deposition (Jessen et al., 2020; Oijen et al., 2007). Therefore, identifying potential beneficial and modifiable lifestyle factors associated with fewer SCCs may provide insight into factors that support cognitive health and may help inform strategies aimed at reducing the risk of later cognitive decline and dementia (Livingston et al., 2020).

Physical activity (PA) is one of the most promising and modifiable lifestyle factors associated with slower cognitive decline and reduced dementia risk (Fratiglioni et al., 2004). Accumulating evidence suggests that AD and related neurodegenerative processes begin decades before the onset of clinical symptoms, highlighting the importance of identifying modifiable exposures during midlife and earlier life stages (Livingston et al., 2020; Norton et al., 2014). Midlife represents a particularly important period during which lifestyle factors, including PA, may influence vascular, metabolic, and neurodegenerative pathways that contribute to later cognitive decline and dementia risk (Fondell et al., 2018; Livingston et al., 2020; Norton et al., 2014). In this context, SCCs provide a potentially sensitive indicator of early cognitive changes that may precede clinically detectable impairment, making them a useful outcome for studying early determinants of cognitive health.

Prior studies examining PA across adulthood suggest that higher levels of PA during early- and mid-adulthood are associated with better later-life objectively-measured cognitive outcomes (Fondell et al., 2018; Kraal et al., 2021; Marino et al., 2025). Focusing on midlife PA may also reduce concerns about reverse causation, as PA in later life may be influenced by prodromal cognitive decline or emerging health limitations (Andrade, 2020). In the Health Professionals Follow-up study, which followed 28,481 men with a mean baseline age of 50 over 20 years, midlife PA was associated with a lower risk of later SCCs (Fondell et al., 2018). However, evidence in women remains limited, despite important sex differences in dementia risk, health behaviors, and lifespan trajectories (Beam et al., 2018; Ferretti et al., 2018; Livingston et al., 2020).

The objective of this study was to address this knowledge gap by examining whether midlife PA is associated with SCCs reported approximately 30 years later in a subset of the New York University Women’s Health Study (NYUWHS). The study adds to the literature by focusing on SCCs, an early and under-studied indicator along the cognitive aging continuum, and by leveraging a prospective cohort of women with detailed lifestyle and health data. In addition, we examines midlife PA, a potentially important exposure window for later-life cognitive health, and uses a validated SCC measure consistent with that used in the Health Professionals Follow-up Study, facilitating comparability across studies. We hypothesized that higher levels of midlife PA would be associated with a lower likelihood of reporting SCCs in later life.

Method

Data Source and Study Design

This prospective nested case-control study was conducted among participants enrolled in the NYUWHS using cumulative incidence sampling (Rothman et al., 2008). Women aged 35 to 65 who visited the Guttman Breast Diagnostic Institute, a large mammography screening clinic in New York City, were invited to participate between 1985 and 1991. Woman who had used hormone therapy or who had been pregnant or nursing within the preceding six months were excluded. A total of 14,274 healthy women were enrolled (Toniolo et al., 1991, 1995). At baseline, participants completed a questionnaire assessing sociodemographic characteristics, health conditions, reproductive history, and lifestyle factors. Over the subsequent 30+ years, nine follow-up rounds were conducted through mail and telephone, and additional questions were added to the surveys to capture features of the aging cohort.

SCCs data were collected in two surveys conducted between 2018 and 2022. Participants completed the questionnaires independently, and response completeness was verified. For participants unable to complete the questionnaires because of cognitive or other health issues, abbreviated questionnaires were completed by relatives; however, these did not include SCC items.

We included participants who completed at least one SCC assessment between 2018 and 2022. Among 6,535 participants who were alive and actively followed at the time of the SCC surveys, 4,397 (67.3%) completed at least one SCC survey (Song et al., 2023).

All study procedures were approved by NYU School of Medicine Institutional Review Board.

Measures

Assessment of Physical Activity

At enrollment, each participant was asked to report the average number of hours spent on PA per week, including vigorous PA (i.e., running, jogging, vigorous bicycling, vigorous swimming, vigorous gym aerobics), moderate PA (i.e., fast walking, doubles tennis, moderate bicycling, moderate swimming, moderate gym aerobics, dancing), and mild PA (i.e., gentle yoga, bowling, golf, light gardening, easy walking). Because PA was assessed only at enrollment, changes in PA over time could not be examined in the present study.

PA were converted to metabolic equivalent of the task (MET), in accordance with earlier studies (Mendes et al., 2018). Vigorous, moderate, and mild PA were assigned values of 9, 4.5, and 3 METs, respectively (Ainsworth et al., 2011). For each intensity level, MET-hours per week (METs-hr/week) were calculated by multiplying the assigned MET value by the number of hours per week spent in that particular activity (Ainsworth et al., 2011; Mendes et al., 2018). Total midlife PA was calculated as the sum of MET-hours per week across vigorous, moderate, and mild activities.

Assessment of Subjective cognitive complaints

In two follow-up surveys conducted between 2018 and 2022, six self-reported yes/no questions were used to evaluate SCCs. Participants were asked whether they experienced the following complaints: 1) general changes in the ability to remember things; 2) difficulty remembering recent events; 3) difficulty remembering a short list of items, such as a shopping list; 4) difficulty understanding or follow spoken instructions; 5) difficulty following a group conversation or a plot in a TV program; 6) difficulty navigating familiar streets (Song et al., 2023; Wu et al., 2022). These questions have demonstrated validity versus objective dementia symptoms (Amariglio et al., 2011). Every complaint was worth 1 point, so the range of SCCs score was 0-6, with higher score indicating more complaints. Consistent with our previous studies (Song et al., 2023; Wu et al., 2022), cases were defined as women with a higher SCCs burden (reporting ⩾2 SCCs), while controls were women with a lower SCCs burden (reporting <2 SCCs).

Very recently, we conducted linkage with the Centers for Medicare & Medicaid Services (CMS) to identify Alzheimer's disease and related dementia cases in this cohort, as described (Walker et al., 2026; Zhang et al., 2025). Briefly, AD/ADRD diagnoses were ascertained using data from the Chronic Conditions segment of the CMS Master Beneficiary Summary File from 1999 to 2021. AD cases were identified according to the CMS algorithm, which requires at least one inpatient, outpatient, carrier, home health, or skilled nursing facility claim with a qualifying diagnosis code during a 3-year look-back period. AD/ADRD cases were identified using a similar CMS algorithm.

Covariates

Covariates were selected a priori based on prior literature and biological plausibility as potential confounders of the association between midlife PA and SCCs. These covariates included demographic, lifestyle, dietary, and health-related factors that are known to be associated with both PA behaviors and cognitive outcomes (Demurtas et al., 2020; Rueda et al., 2025; Zhong et al., 2025). Data on demographic and lifestyle variables were collected at baseline, including age (continuous, baseline), menopausal status (premenopausal or postmenopausal), race/ethnicity (White, African-American/ Latina, or Asian/ Native American/ more than one race), educational attainment (< college, college, or ≥ graduate), smoking status (never or ever), Grams ethanol per week, and body mass Index (underweight, normal, overweight, or obesity). Dietary Approaches to Stop Hypertension score (DASH score; 7-17, 18-20, 21-24, or 25-35) was calculated based on a validated food frequency questionnaire. Other covariates related to health conditions and medical history were collected on follow-up surveys, including history of depression (yes or no), cardiometabolic disease (yes or no; including hypertension, diabetes, stroke, and coronary heart disease), and cancer, which was determined based on self-report, with confirmation by review of medical records, as well as linkages to tumor registries of New York, New Jersey and Florida.

Statistical analysis

Baseline demographic and clinical characteristics were then compared between SCCs groups. Continuous variables were summarized as means ± standard deviations (SD) and categorical variables as frequencies (%). Group differences were assessed using t-tests for continuous variables and chi-square tests for categorical variables.

To assess the association between PA and SCCs, we classified the total PA and its components by tertile and computed the OR (95% CI) for SCCs using unconditional logistic regression comparing women in each of the higher two tertiles with those in the bottom tertile of PA. Missing data (0.3%-7.5% for each covariate) were included in the model using dummy variables. We adjusted for demographics and risk factors for SCCs in the first model, including age, menopausal status, race, education, smoking status, and BMI. We then added more covariates in model 2, i.e., alcohol consumption, history of cancer, depression, cardiometabolic disease, and the DASH score. We also assessed the linear trend across tertiles by computing the p-trend and the OR (95% CI) per tertile.

We conducted a stratified analysis to assess potential effect modifiers including age (<44.9, ≥44.9; by median), race (White, other), educational attainment (< college, ≥ college), smoking status (ever, never), alcohol consumption (0, >0 gram per week), and DASH diet scale (<21, ≥21; by median). We tested multiplicative interactions using cross-product terms between the top two PA tertiles and each dichotomized potential effect modifier, with p-values indicating significance. Given our prior finding of an inverse association between DASH adherence and SCCs (Song et al., 2023), we also assessed the joint effects of DASH and PA on SCCs.

As part of our sensitivity analysis, we employed a multiple imputation (MI) model based on Markov chain Monte Carlo (MCMC) (Austin et al., 2021; Javanbakht et al., 2022). In this MI model, we imputed missing data for baseline demographic variables such as age, menopausal status, race, educational attainment, BMI, and smoking status. We also conducted an inverse probability weighting (IPW) model to address potential selection bias and improve the precision and generalization of the estimated association (Sayon-Orea et al., 2020; Tang et al., 2020). Another sensitivity analyses were conducted excluding 497 women with a diagnosis of AD/ADRD before SCC assessment, based on CMS data.

We conducted exploratory analyses using alternative outcome definitions to evaluate the robustness of findings. To improve measurement reliability across the 2-to-3-year interval between questionnaires, SCC data from women who completed both assessments (n=3,240) were combined, and a total SCC score (range: 0–12) was derived by summing responses across the two surveys. The resulting SCC score was categorized as none (0 complaints; 43.7%), moderate (1–2 complaints; 29.2%), and high (>2 complaints; 27.1%). Multinomial logistic regression models were used to examine the associations between PA tertiles and SCCs levels, using no SCC as the reference group. All analyses were conducted using SAS version 9.4 and RStudio Version 4.2.2.

Results

This study involved 4,397 women, 32.4% of whom reported experiencing ≥ 2 SCCs in at least one survey. The distribution of SCC scores (0–6) was as follows: 1920 women (43.67%) reported 0 SCCs, 1054 (23.97%) reported 1 SCC, 666 (15.15%) reported 2 SCCs, 426 (9.69%) reported 3 SCCs, 152 (3.46%) reported 4 SCCs, 94 (2.14%) reported 5 SCCs, and 85 (1.93%) reported 6 SCCs. The mean age was 46.2 years at baseline, the mean follow-up time was 32.6 years (range 27.0-37.0), and the mean age at completion of the SCCs questionnaire was 78.7 years (range 62.5-99.7) (Table 1). Compared with women with <2 SCCs, those who reported ≥2 SCCs were older (p<0.001), had lower educational attainment (p=0.007), were more likely to be ever smokers (p=0.003), and had a higher BMI (p=0.029) at enrollment. They also had significantly lower moderate PA at enrollment (p=0.007). At the time of SCCs assessment, they were more likely to be older (p<0.001), have a history of cardiometabolic disease (p<0.001), and/or a history of depression (p<0.001).

Table 1.

Distribution of sociodemographic and health-related characteristics by subjective cognitive complaints (≥ 2 versus < 2), xx city, 1989-2022

Total
(N = 4397)		 Subjective cognitive complaints
 p-valuea
SCCs < 2
(n = 2974)		 SCCs ≥ 2
(n = 1423)
Demographic variables at baseline
Age, mean (std), years 46.16 (7.11) 45.62 (6.84) 47.29 (7.53) <0.001b
Menopausal status, n (%) <0.001
 Premenopausal 3290 (74.82) 2303 (77.44) 987 (69.36)
 Postmenopausal 1107 (25.18) 671 (22.56) 436 (30.64)
Race/ethnicity1, n (%) 0.104
 White 3481 (82.45) 2369 (83.30) 1112 (80.70)
 African-American/ Latina 673 (15.94) 433 (15.23) 240 (15.94)
 Asian/ Native American/ More than one race 68 (1.61) 42 (1.48) 26 (1.89)
Smoking2, n (%) 0.003
 Never 2051 (46.92) 1431 (48.49) 620 (43.66)
 Ever 2320 (53.08) 1520 (51.51) 800 (56.34)
Education3, n (%) 0.007
 < College 1723 (42.38) 1117 (40.74) 606 (45.77)
 College 1070 (26.32) 733 (26.73) 337 (25.45)
 Graduate school or more 1273 (31.31) 892 (32.53) 381 (28.78)
Grams ethanol per week4, median (IQR), g 42.92 (84.32) 41.23 (83.46) 46.49 (86.03) 0.057b
BMI5c, n(%), kg/m2 0.029
 Underweight 95 (2.17) 73 (2.47) 22 (1.55)
 Normal 2884 (65.97) 1970 (66.62) 914 (64.59)
 Overweight 979 (22.39) 654 (22.12) 325 (22.97)
 Obesity 414 (9.47) 260 (8.79) 154 (10.88)
DASH score6d, n (%) 0.145
 Q1 (7-17) 1094 (25.18) 724 (24.62) 370 (26.18)
 Q2 (18-20) 944 (21.73) 636 (21.63) 308 (21.95)
 Q3 (21-24) 1249 (28.75) 877 (29.82) 372 (26.51)
 Q4 (25-35) 1057 (24.33) 704 (23.94) 353 (25.16)
Follow-up variables at the time of SCD
Age, mean (std), years 78.74 (7.55) 78.13 (7.24) 80.02 (8.00) <0.001b
History of cancer, n (%) 1499 (34.09) 987 (33.19) 512 (35.98) 0.068
History of depression1, n (%) 832 (18.99) 441 (14.87) 391 (27.61) <0.001
History of cardiometabolic diseasee, n (%) 3013 (68.52) 1977 (66.48) 1036 (72.80) <0.001
Midlife physical activity, MET-hours per week
Totalf, mean (std) 45.12 (45.91) 45.94 (45.21) 43.42 (47.30) 0.093b
 Vigorous exercise, mean (std) 18.75 (26.71) 19.14 (26.02) 17.94 (28.08) 0.172b
 Moderate exercise, mean (std) 13.47 (18.41) 13.97 (18.98) 12.42 (17.12) 0.007b
 Mild exercise, mean (std) 12.90 (14.18) 12.82 (13.63) 13.07 (15.28) 0.610b
Subjective cognitive Complaints
Number
 0, n (%) 1920 (43.67)
 1, n (%) 1054 (23.97)
 2, n (%) 666 (15.15)
 3, n (%) 426 (9.69)
 4, n (%) 152 (3.46)
 5, n (%) 94 (2.14)
 6, n (%) 85 (1.93)
Open in a new tab
1

Missing = 175

2

Missing = 26

3

Missing = 331

4

Missing = 133

5

Missing = 25

6

Missing = 53

a

Pearson’s χ2 test unless otherwise indicated

b

t-Test for independent sample

c

Body Mass Index, estimates were calculated from self-reported weight and height. Underweight indicates BMI < 18.5, normal indicates 18.5 ≤ BMI < 25, overweight indicates 25 ≤ BMI < 30, and obesity indicates BMI ≥ 30.

d

Dietary Approaches to Stop Hypertension score, ranged from 7 to 35, with a higher score indicating a healthier diet. DASH score was divided into quartiles based on the following ranges: Q1, 7-17; Q2, 18-20; Q3, 21-24; and Q4, 25-35.

e

Cardiometabolic disease includes hypertension, diabetes, stroke, and coronary heart disease.

f

Total physical activity (MET-hours/week) includes vigorous MPA (9 MET), moderate MPA (4.5 MET), and mild MPA (3 MET).

1

Missing = 15

Women in the highest tertile of total midlife PA had a 19% reduction in the odds of having ≥2 SCCs, compared to those in the bottom tertile, after adjusting for age, menopausal status, race, educational attainment, smoking status, and BMI (OR=0.81; 95% CI: 0.69-0.94; p-trend=0.008; model 1, Table 2). The association remained after adjustment for additional covariates (OR=0.80; 95% CI: 0.68-0.95; p-trend=0.009; model 2). Every tertile difference in PA was associated with an OR of 0.90 (95% CI: 0.83-0.97) for ≥2 SCCs in model 1 and 0.89 (95% CI: 0.82-0.97) in model 2. Among different types of PA, moderate PA was significantly inversely associated with odds of having ≥2 SCCs (p-trend for model 1 = 0.020; for model 2 = 0.041), while the association for vigorous PA was consistent but not statistically significant.

Table 2.

Associations between midlife physical activity (MET hours per week) and subjective cognitive complaints (≥ 2 versus < 2)

Tertile of Midlife Physical Activity
OR (95%CI)
 p-trend OR (95%CI) per tertile
T1 T2 T3
Total PA
 Median [range] 6.0 [0-20.7] 34.0 [20.9-51.1] 79.0 [51.2-550.0]
 Cases/Controls 515/951 476/989 432/1034
 Model 1a Ref 0.94 (0.80-1.11) 0.81 (0.69-0.95) 0.008 0.90 (0.83-0.97)
 Model 2b Ref 0.96 (0.81-1.11) 0.80 (0.68-0.95) 0.009 0.89 (0.82-0.97)
Vigorous PA
 Median [range] 0 [0-0.2] 9.6 [0.3-21.8] 36.0 [22.0-377.5]
 Cases/Controls 506/959 472/989 445/1026
 Model 1 Ref 0.98 (0.83-1.15) 0.88 (0.75-1.03) 0.109 0.94 (0.87-1.02)
 Model 2 Ref 1.01 (0.85-1.19) 0.87 (0.74-1.02) 0.111 0.93 (0.86-1.02)
Moderate PA
 Median [range] 0 [0-0] 9.0 [4.5-13.5] 27.0 [18.0-324.0]
 Cases/Controls 620/1180 365/753 438/1041
 Model 1 Ref 1.01 (0.86-1.20) 0.83 (0.71-0.97) 0.020 0.91 (0.85-0.99)
 Model 2 Ref 1.03 (0.87-1.22) 0.84 (0.71-0.99) 0.041 0.92 (0.85-1.00)
Mild PA
 Median [range] 0 [0-3.0] 9.0 [6.0-12.0] 21.0 [15.0-150.0]
 Cases/Controls 427/876 432/924 564/1174
 Model 1 Ref 1.00 (0.85-1.19) 1.02 (0.87-1.19) 0.839 1.01 (0.93-1.09)
 Model 2 Ref 1.00 (0.84-1.18) 1.03 (0.88-1.21) 0.683 1.02 (0.94-1.10)
Open in a new tab
a

Model 1 was adjusted for age (continuous, baseline), menopausal status (premenopausal or postmenopausal), race (White, African-American, Latina, or other (Asian, Native American, more than one race)), education (< college, college, or ≥ graduate), smoking status (never or ever), Body Mass Index (underweight, normal, overweight, or obesity).

b

Model 2 had the same adjustments as model 1 plus alcohol consumption, history of cancer (yes or no), history of depression (yes or no), history of cardiometabolic disease (yes or no; including hypertension, diabetes, stroke, and heart disease), DASH score (7-17, 18-20, 21-24, or 25-35).

In stratified analyses (Figure 1.A), the association was stronger among never smokers compared to ever smokers (p-value for interaction = 0.029). The associations between the highest tertile of total PA and having ≥2 SCCs did not differ by other potential effect modifiers. When considering the combined effect between DASH and PA, significant 28% reduced odds of SCCs was observed in the high adherence to the DASH diet and highest tertile of PA compared to the low DASH and lowest tertile of PA (OR=0.72; 95% CI: 0.58-0.89; p=0.002; Figure 1.B).

Figure 1. Subgroup analyses of the associations between total midlife physical activity (MET hours per week) and subjective cognitive complaints (≥ 2 versus < 2).

Figure 1.

Figure 1.

Open in a new tab

a Subjects with missing values for a particular stratifying variable were excluded from analyses stratified by that variable but were included in analyses stratified by other variables. Analyses stratified by race, smoking, education, alcohol consumption, and DASH excluded 175, 26, 331, 133, 53 subjects, respectively.

b Model was adjusted for age (continuous, baseline), menopausal status (premenopausal or postmenopausal), race (White, African-American, Latina, or other (Asian, Native American, more than one race)), education (< college, college, or ≥ graduate), smoking status (never or ever), Body Mass Index (underweight, normal, overweight, or obesity), alcohol consumption, history of cancer (yes or no), history of depression (yes or no), history of cardiometabolic disease (yes or no; including hypertension, diabetes, stroke, and heart disease), and DASH score (7-17, 18-20, 21-24, or 25-35)

Model was adjusted for age (continuous, baseline), menopausal status (premenopausal or postmenopausal), race (White, African-American, Latina, or other (Asian, Native American, more than one race)), education (< college, college, or ≥ graduate), smoking status (never or ever), Body Mass Index (underweight, normal, overweight, or obesity), alcohol consumption, history of cancer (yes or no), history of depression (yes or no), and history of cardiometabolic disease (yes or no; including hypertension, diabetes, stroke, and heart disease)

In sensitivity analysis, a similar pattern of inverse associations between the highest tertile of total PA and SCCs was observed when imputing missing values or using IPW (Supplemental Table 1). Findings were also similar after excluding 497 women diagnosed with AD/ADRD before SCC assessments (Supplemental Table 2). In the analyses using 3 categories of SCCs, the highest tertile of total PA as compared to the lowest tertile was associated with 22% lower odds of a poor SCCs score (SCCs > 2) (OR=0.78; 95% CI: 0.63-0.98; p-trend=0.031; Supplemental Table 3).

Discussion

This study suggested that engaging in high levels of PA during midlife (average age of 46.2 years) was associated with a 20% reduction in the odds of reporting ≥2 SCCs after 30 years. The association was more pronounced in participants who had never smoked. The combined effects between PA and DASH adherence were additive. The direction of association was consistent for both vigorous and moderate types of PA but not mild PA.

Previous studies reported inconsistent results regarding the relationship between PA and objective cognitive function. In the NHANES study of 2,336 participants over 60 years old, compared with the inactive group, high and moderate intensity activity was associated with good cognitive performance, respectively (Li et al., 2023). In the Nurses’ Health Study (n=18,766 women aged 70–81), average PA over 8–15 years was associated with a 20% lower risk of objective cognitive decline for those in the highest versus lowest quintile of total PA (Weuve, 2004). Another NHS study of 7,252 women aged ≥70 found that vigorous PA, assessed at least nine years before cognitive testing, was associated with better cognitive trajectories (Fassier et al., 2022). In the Framingham Study (n=3,714; mean age 70), participants with low PA had a 50% higher risk of dementia over 10 years compared with those with higher PA levels (Tan et al., 2016). In contrast, longitudinal studies from the U.S. (n=1,718; mean baseline age 45.7 years; ~10-year follow-up) (Greendale et al., 2021) and the Netherlands (n=7,176; baseline age >55 years; 9-year follow-up) (Stringa et al., 2020) found no association between PA and cognitive function. The inconsistent findings may reflect differences in the timing of PA assessment, as older adults often reduce activity with age and most studies had follow-ups under 10 years (Cunningham et al., 2020; McPhee et al., 2016; World Health Organization, 2022). The Health Professionals Follow-up study (HPFS) examined the association between PA (average baseline age of 51 years) and subjective cognitive function in mid-to-late life, approximately 20 years later (average age of 75 years) (Fondell et al., 2018). Data on PA was collected repeatedly from midlife to late-life, revealing that higher PA during early or early-to-mid adulthood was associated with a 23% and 48% reduction in poor cognitive function odds, respectively (Fondell et al., 2018). Compared with the HPFS, our participants in NYUWHS were younger at baseline (average age of 46.2 years), and were followed up for a longer period of time (>30 years). The findings align with those from the HPFS study and extend the evidence to women, supporting an association between regular PA and fewer early cognitive symptoms, such as SCCs, before clinical impairment develops.

The findings align well with the potential biological mechanisms. PA regulates blood flow and delivery of oxygen and nutrients to the brain, thus reducing inflammation and oxidative stress, both of which are associated with cognitive decline (Claassen et al., 2021). Moderate-intensity aerobic exercise could increase the production of neurotrophins, such as brain-derived neurotrophic factor (BDNF), which can promote the growth and survival of new neurons in the hippocampus, a key brain region for memory and learning (Khoury et al., 2023; Lippi et al., 2020; Ribeiro et al., 2021). Normal aging in mid-life is associated with a reduction in both structure and function of the brain (Gu et al., 2020; Khoury et al., 2023; Ly et al., 2014). PA may positively influence several pathways that contribute to maintenance of brain health into late life (Macpherson et al., 2017). One recent study observed a positive association between self-reported leisure time PA in middle-aged adults and larger total brain volume in elderly, specifically attributed to increased cerebral gray matter volume, after adjustment for confounders (Rovio et al., 2010). The strong and significant association between moderate PA and reduced SCCs, compared to vigorous PA, may be attributed to differences in energy metabolism. Moderate PA predominantly utilizes fat oxidation, while vigorous activity relies more on carbohydrate metabolism (Melzer, 2011). Fat oxidation during moderate exercise has been linked to long-term health benefits, including improved cardiovascular health and enhanced metabolic efficiency, which may have downstream effects on cognitive function (Achten & Jeukendrup, 2004). This conditioning could be particularly advantageous for individuals whose primary goal is health maintenance rather than performance. Future research should investigate these physiological mechanisms further to clarify why moderate PA might offer greater cognitive protection than vigorous PA.

This study has several strengths, including its prospective design, large sample size at recruitment, and extended follow-up time of over 30 years. Data on PA were collected 30 years before the outcome assessment, reducing the likelihood of reverse causation. The findings were robust with multiple analytic methods and sensitivity analyses. However, there are some limitations to consider. First, although the study population consisted of 20% non-White, we observed that the association between midlife PA and SCCs was generally consistent across key subgroups, including age and race/ethnicity. In addition, sensitivity analyses using IPW produced similar results, suggesting that potential selection bias due to attrition or survival did not materially influence the observed associations. Second, although SCCs were measured using a self-reported questionnaire, which may introduce measurement error, the SCC questions have been validated and shown to have long-term reproducibility in other populations (Fondell et al., 2018; Oijen et al., 2007; Omura et al., 2020; Samieri et al., 2014). Recent data indicate that that SCCs may be linked to pathological changes of AD (Grill et al., 2015), such as amyloid-β (Aβ) deposition (Amariglio et al., 2012; Schwarz et al., 2021), increases in white matter lesions (De Groot et al., 2001; Dhana et al., 2022), and gray matter atrophy (Peter et al., 2014). Similarly, PA was assessed only once at baseline through self-reported measures, which may not fully capture long-term patterns of activity. However, a single assessment of PA has been shown to be reliable and valid measure for long-term PA and has been associated with the risk of chronic diseases later in life in prospective epidemiologic cohort studies (Cust et al., 2008; Hu et al., 1999; Manson et al., 1999).We also acknowledge that although this study was adjusted for a range of demographic, lifestyle, dietary, and health-related factors, such as BMI, diet habits, and history of various chronic diseases that selected a priori, like other epidemiologic studies, residual confounding by unmeasured or incompletely measured factors remains possible. In addition, the SCC assessments partly overlapped with the COVID-19 pandemic. Given that the COVID-19 pandemic occurred decades after the assessment of midlife PA at baseline (1985–1991), it is unlikely to have confounded the observed association. However, pandemic-related factors, including social networks, stress, and healthcare disruption may have influenced cognitive functioning or participants’ perceptions and reporting of cognitive difficulties (Kassam & McMillan, 2023; Zsuffa et al., 2024). Consequently, the findings should be interpreted with this potential influence on SCC assessment in mind. Future studies should examine whether later-life factors, such as social isolation from the pandemic, can modify the association between midlife PA and cognitive health. Finally, although data on repeated measurement of SCCs were available for 3240 women and were used to improve the reliability of SCC classification, we did not examine longitudinal changes in SCCs because the duration between the two SCCs was relatively short (2-3 years). The literature suggests that individuals with SCCs are at increased risk of developing dementia (Jessen et al., 2014); however, no established threshold for the number of SCCs has been identified. SCCs should be interpreted as general cognitive concerns and potentially early markers of cognitive vulnerability. In this study, sensitivity analyses excluding women with AD/ADRD that suggest that PA might be related to SCCs independently of diagnosed neurogenerative disease. Future studies incorporating longitudinal objective cognitive assessments and SCCs are needed to clarify the clinical implications of SCCs.

Conclusions and Implications

In conclusion, an inverse association was found between PA during midlife and SCCs in late life. These findings suggest that regular midlife PA may be a potential protective factor for later-life cognitive health among women. Further research using objective measures of cognitive function is needed to investigate effects of different types of PA at different stages of life.

Supplementary Material

Supplementary Material

Acknowledgements:

The authors have no acknowledgments to report.

Footnotes

Conflicts of Interest: All authors have read the manuscript, agreed that the work is ready for submission to a journal, and declare no conflicts of interest related to this research.

REFERENCE

  1. Achten J, & Jeukendrup AE. (2004). Optimizing fat oxidation through exercise and diet. Nutrition, 20(7–8), 716–727. 10.1016/j.nut.2004.04.005 [DOI] [PubMed] [Google Scholar]
  2. Ainsworth BE, Haskell WL, Herrmann SD, Meckes N, Bassett DR, Tudor-Locke C, Greer JL, Vezina J, Whitt-Glover MC, & Leon AS. (2011). 2011 Compendium of Physical Activities: A Second Update of Codes and MET Values. Medicine & Science in Sports & Exercise, 43(8), 1575–1581. 10.1249/MSS.0b013e31821ece12 [DOI] [PubMed] [Google Scholar]
  3. Amariglio RE, Becker JA, Carmasin J, Wadsworth LP, Lorius N, Sullivan C, Maye JE, Gidicsin C, Pepin LC, Sperling RA, Johnson KA, & Rentz DM. (2012). Subjective cognitive complaints and amyloid burden in cognitively normal older individuals. Neuropsychologia, 50(12), 2880–2886. 10.1016/j.neuropsychologia.2012.08.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Amariglio RE, Townsend MK, Grodstein F, Sperling RA, & Rentz DM. (2011). Specific Subjective Memory Complaints in Older Persons May Indicate Poor Cognitive Function: MEMORY COMPLAINTS AND OBJECTIVE COGNITIVE TESTING. Journal of the American Geriatrics Society, 59(9), 1612–1617. 10.1111/j.1532-5415.2011.03543.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Andrade C. (2020). Reverse Causation, Physical Inactivity, and Dementia. Indian Journal of Psychological Medicine, 42(2), 205–206. 10.4103/IJPSYM.IJPSYM_45_20 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Austin PC, White IR, Lee DS, & van Buuren S. (2021). Missing Data in Clinical Research: A Tutorial on Multiple Imputation. Canadian Journal of Cardiology, 37(9), 1322–1331. 10.1016/j.cjca.2020.11.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Beam CR, Kaneshiro C, Jang JY, Reynolds CA, Pedersen NL, & Gatz M. (2018). Differences Between Women and Men in Incidence Rates of Dementia and Alzheimer’s Disease. Journal of Alzheimer’s Disease, 64(4), 1077–1083. 10.3233/JAD-180141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Claassen JAHR, Thijssen DHJ, Panerai RB, & Faraci FM. (2021). Regulation of cerebral blood flow in humans: Physiology and clinical implications of autoregulation. Physiological Reviews, 101(4), 1487–1559. 10.1152/physrev.00022.2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cunningham C, O’ Sullivan R, Caserotti P, & Tully MA. (2020). Consequences of physical inactivity in older adults: A systematic review of reviews and meta-analyses. Scandinavian Journal of Medicine & Science in Sports, 30(5), 816–827. 10.1111/sms.13616 [DOI] [PubMed] [Google Scholar]
  10. Cust AE, Smith BJ, Chau J, Van Der Ploeg HP, Friedenreich CM, Armstrong BK, & Bauman A. (2008). Validity and repeatability of the EPIC physical activity questionnaire: A validation study using accelerometers as an objective measure. International Journal of Behavioral Nutrition and Physical Activity, 5(1), 33. 10.1186/1479-5868-5-33 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. De Groot JC, De Leeuw F-E, Oudkerk M, Hofman A, Jolles J, & Breteler MMB. (2001). Cerebral white matter lesions and subjective cognitive dysfunction: The Rotterdam Scan Study. Neurology, 56(11), 1539–1545. 10.1212/WNL.56.11.1539 [DOI] [PubMed] [Google Scholar]
  12. Demurtas J, Schoene D, Torbahn G, Marengoni A, Grande G, Zou L, Petrovic M, Maggi S, Cesari M, Lamb S, Soysal P, Kemmler W, Sieber C, Mueller C, Shenkin SD, Schwingshackl L, Smith L, Veronese N, & European Society of Geriatric Medicine Special Interest Group in Systematic Reviews and Meta-Analyses, Frailty, Sarcopenia, and Dementia. (2020). Physical Activity and Exercise in Mild Cognitive Impairment and Dementia: An Umbrella Review of Intervention and Observational Studies. Journal of the American Medical Directors Association, 21(10), 1415–1422.e6. 10.1016/j.jamda.2020.08.031 [DOI] [PubMed] [Google Scholar]
  13. Dhana A, DeCarli C, Dhana K, Desai P, Krueger K, Evans DA, & Rajan KB. (2022). Association of Subjective Memory Complaints With White Matter Hyperintensities and Cognitive Decline Among Older Adults in Chicago, Illinois. JAMA Network Open, 5(4), e227512. 10.1001/jamanetworkopen.2022.7512 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fassier P, Kang JH, Lee I-M, Grodstein F, & Vercambre M-N. (2022). Vigorous Physical Activity and Cognitive Trajectory Later in Life: Prospective Association and Interaction by Apolipoprotein E e4 in the Nurses’ Health Study. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 77(4), 817–825. 10.1093/gerona/glab169 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ferretti MT, Iulita MF, Cavedo E, Chiesa PA, Schumacher Dimech A, Santuccione Chadha A, Baracchi F, Girouard H, Misoch S, Giacobini E, Depypere H, Hampel H, & Women’s Brain Project and the Alzheimer Precision Medicine Initiative. (2018). Sex differences in Alzheimer disease—The gateway to precision medicine. Nature Reviews. Neurology, 14(8), 457–469. 10.1038/s41582-018-0032-9 [DOI] [PubMed] [Google Scholar]
  16. Fondell E, Townsend MK, Unger LD, Okereke OI, Grodstein F, Ascherio A, & Willett WC. (2018). Physical activity across adulthood and subjective cognitive function in older men. European Journal of Epidemiology, 33(1), 79–87. 10.1007/s10654-017-0331-2 [DOI] [PubMed] [Google Scholar]
  17. Fratiglioni L, Paillard-Borg S, & Winblad B. (2004). An active and socially integrated lifestyle in late life might protect against dementia. The Lancet Neurology, 3(6), 343–353. 10.1016/S1474-4422(04)00767-7 [DOI] [PubMed] [Google Scholar]
  18. Greendale GA, Han W, Huang M, Upchurch DM, Karvonen-Gutierrez C, Avis NE, & Karlamangla AS. (2021). Longitudinal Assessment of Physical Activity and Cognitive Outcomes Among Women at Midlife. JAMA Network Open, 4(3), e213227. 10.1001/jamanetworkopen.2021.3227 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Grill JD, Vinters HV, & Monsell SE. (2015). Does Alzheimer Disease Pathologic Change Underlie Subjective Cognitive Complaints? Alzheimer Disease & Associated Disorders, 29(4), 350–352. 10.1097/WAD.0000000000000092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gu Y, Beato JM, Amarante E, Chesebro AG, Manly JJ, Schupf N, Mayeux RP, & Brickman AM. (2020). Assessment of Leisure Time Physical Activity and Brain Health in a Multiethnic Cohort of Older Adults. JAMA Network Open, 3(11), e2026506. 10.1001/jamanetworkopen.2020.26506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hu FB, Sigal RJ, Rich-Edwards JW, Colditz GA, Solomon CG, Willett WC, Speizer FE, & Manson JE. (1999). Walking Compared With Vigorous Physical Activity and Risk of Type 2 Diabetes in Women: A Prospective Study. JAMA, 282(15), 1433. 10.1001/jama.282.15.1433 [DOI] [PubMed] [Google Scholar]
  22. Javanbakht M, Lin J, Ragsdale A, Kim S, Siminski S, & Gorbach P. (2022). Comparing single and multiple imputation strategies for harmonizing substance use data across HIV-related cohort studies. BMC Medical Research Methodology, 22(1), 90. 10.1186/s12874-022-01554-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jessen F, Amariglio RE, Boxtel M, Breteler M, Ceccaldi M, Chételat G, Dubois B, Dufouil C, Ellis KA, Flier WM, Glodzik L, Harten AC, Leon MJ, McHugh P, Mielke MM, Molinuevo JL, Mosconi L, Osorio RS, Perrotin A, … Subjective Cognitive Decline Initiative (SCD-I) Working Group. (2014). A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease. Alzheimer’s & Dementia, 10(6), 844–852. 10.1016/j.jalz.2014.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jessen F, Amariglio RE, Buckley RF, Van Der Flier WM, Han Y, Molinuevo JL, Rabin L, Rentz DM, Rodriguez-Gomez O, Saykin AJ, Sikkes SAM, Smart CM, Wolfsgruber S, & Wagner M. (2020). The characterisation of subjective cognitive decline. The Lancet Neurology, 19(3), 271–278. 10.1016/S1474-4422(19)30368-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kassam K, & McMillan JM. (2023). The impact of loneliness and social isolation during COVID-19 on cognition in older adults: A scoping review. Frontiers in Psychiatry, 14, 1287391. 10.3389/fpsyt.2023.1287391 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Khoury R, Saad J, Jabre V, Ghayad LM, Khalifeh M, Houbeika R, El Ahmad P, Mezher A, El Masri D, Haddad Z, Eid F, Barmo N, Nasrallah P, Sleiman SF, & Stephan JS. (2023). Autophagy regulates the release of exercise factors and their beneficial effects on spatial memory recall. Heliyon, 9(4), e14705. 10.1016/j.heliyon.2023.e14705 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kraal AZ, Dotterer HL, Sharifian N, Morris EP, Sol K, Zaheed AB, Smith J, & Zahodne LB. (2021). Physical Activity in Early- and Mid-Adulthood Are Independently Associated With Longitudinal Memory Trajectories in Later Life. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 76(8), 1495–1503. 10.1093/gerona/glaa252 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Li W, Li Y, Wen M, & Liu B. (2023). Physical Activity Patterns and Cognitive Health among Older Adults in the United States. Journal of Applied Gerontology, 42(3), 409–418. 10.1177/07334648221139480 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lippi G, Mattiuzzi C, & Sanchis-Gomar F. (2020). Updated overview on interplay between physical exercise, neurotrophins, and cognitive function in humans. Journal of Sport and Health Science, 9(1), 74–81. 10.1016/j.jshs.2019.07.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, Brayne C, Burns A, Cohen-Mansfield J, Cooper C, Costafreda SG, Dias A, Fox N, Gitlin LN, Howard R, Kales HC, Kivimäki M, Larson EB, Ogunniyi A, … Mukadam N. (2020). Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The Lancet, 396(10248), 413–446. 10.1016/S0140-6736(20)30367-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Ly M, Canu E, Xu G, Oh J, McLaren DG, Dowling NM, Alexander AL, Sager MA, Johnson SC, & Bendlin BB. (2014). Midlife measurements of white matter microstructure predict subsequent regional white matter atrophy in healthy adults. Human Brain Mapping, 35(5), 2044–2054. 10.1002/hbm.22311 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Macpherson H, Teo W-P, Schneider LA, & Smith AE. (2017). A Life-Long Approach to Physical Activity for Brain Health. Frontiers in Aging Neuroscience, 9, 147. 10.3389/fnagi.2017.00147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Manson JE, Hu FB, Rich-Edwards JW, Colditz GA, Stampfer MJ, Willett WC, Speizer FE, & Hennekens CH. (1999). A Prospective Study of Walking as Compared with Vigorous Exercise in the Prevention of Coronary Heart Disease in Women. New England Journal of Medicine, 341(9), 650–658. 10.1056/NEJM199908263410904 [DOI] [PubMed] [Google Scholar]
  34. Marino FR, Lyu C, Li Y, Liu T, Au R, & Hwang PH. (2025). Physical Activity Over the Adult Life Course and Risk of Dementia in the Framingham Heart Study. JAMA Network Open, 8(11), e2544439. 10.1001/jamanetworkopen.2025.44439 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McPhee JS, French DP, Jackson D, Nazroo J, Pendleton N, & Degens H. (2016). Physical activity in older age: Perspectives for healthy ageing and frailty. Biogerontology, 17(3), 567–580. 10.1007/s10522-016-9641-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Melzer K. (2011). Carbohydrate and fat utilization during rest and physical activity. E-SPEN, the European e-Journal of Clinical Nutrition and Metabolism, 6(2), e45–e52. 10.1016/j.eclnm.2011.01.005 [DOI] [Google Scholar]
  37. Mendes M. de A., da Silva I, Ramires V, Reichert F, Martins R, Ferreira R, & Tomasi E, (2018). Metabolic equivalent of task (METs) thresholds as an indicator of physical activity intensity. PLOS ONE, 13(7), e0200701. 10.1371/journal.pone.0200701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Norton S, Matthews FE, Barnes DE, Yaffe K, & Brayne C. (2014). Potential for primary prevention of Alzheimer’s disease: An analysis of population-based data. The Lancet Neurology, 13(8), 788–794. 10.1016/S1474-4422(14)70136-X [DOI] [PubMed] [Google Scholar]
  39. Oijen M, Jong FJ, Hofman A, Koudstaal PJ, & Breteler MMB. (2007). Subjective memory complaints, education, and risk of Alzheimer’s disease. Alzheimer’s & Dementia, 3(2), 92–97. 10.1016/j.jalz.2007.01.011 [DOI] [PubMed] [Google Scholar]
  40. Omura JD, Brown DR, McGuire LC, Taylor CA, Fulton JE, & Carlson SA. (2020). Cross-sectional association between physical activity level and subjective cognitive decline among US adults aged ≥45 years, 2015. Preventive Medicine, 141, 106279. 10.1016/j.ypmed.2020.106279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Peter J, Scheef L, Abdulkadir A, Boecker H, Heneka M, Wagner M, Koppara A, Klöppel S, Jessen F, & Alzheimer’s Disease Neuroimaging Initiative. (2014). Gray matter atrophy pattern in elderly with subjective memory impairment. Alzheimer’s & Dementia, 10(1), 99–108. 10.1016/j.jalz.2013.05.1764 [DOI] [PubMed] [Google Scholar]
  42. Prince M, Ali G-C, Guerchet M, Prina AM, Albanese E, & Wu Y-T. (2016). Recent global trends in the prevalence and incidence of dementia, and survival with dementia. Alzheimer’s Research & Therapy, 8(1), 23. 10.1186/s13195-016-0188-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ribeiro D, Petrigna L, Pereira FC, Muscella A, Bianco A, & Tavares P. (2021). The Impact of Physical Exercise on the Circulating Levels of BDNF and NT 4/5: A Review. International Journal of Molecular Sciences, 22(16), 8814. 10.3390/ijms22168814 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rothman KJ, Greenland S, & Lash TL. (2008). Modern epidemiology (3rd ed., thoroughly rev. and updated). Wolters Kluwer Health/Lippincott Williams & Wilkins. [Google Scholar]
  45. Rovio S, Spulber G, Nieminen LJ, Niskanen E, Winblad B, Tuomilehto J, Nissinen A, Soininen H, & Kivipelto M. (2010). The effect of midlife physical activity on structural brain changes in the elderly. Neurobiology of Aging, 31(11), 1927–1936. 10.1016/j.neurobiolaging.2008.10.007 [DOI] [PubMed] [Google Scholar]
  46. Rueda AV, Sánchez MW, Fernández-Ballbé Ó, De La Torre L, & Sánchez SW. (2025). A comparative analysis of physical activity and cognitive performance in both institutionalized and non-institutionalized older adults, controlling sociodemographic variables. Frontiers in Psychology, 16, 1719468. 10.3389/fpsyg.2025.1719468 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Samieri C, Proust-Lima CM, Glymour M, Okereke OI, Amariglio RE, Sperling RA, Rentz DM, & Grodstein F, (2014). Subjective cognitive concerns, episodic memory, and the APOE ε4 allele. Alzheimer’s & Dementia, 10(6), 752. 10.1016/j.jalz.2014.06.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sayon-Orea C, Moreno-Iribas C, Delfrade J, Sanchez-Echenique M, Amiano P, Ardanaz E, Gorricho J, Basterra G, Nuin M, & Guevara M. (2020). Inverse-probability weighting and multiple imputation for evaluating selection bias in the estimation of childhood obesity prevalence using data from electronic health records. BMC Medical Informatics and Decision Making, 20(1), 9. 10.1186/s12911-020-1020-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Schwarz C, Lange C, Benson GS, Horn N, Wurdack K, Lukas M, Buchert R, Wirth M, & Flöel A. (2021). Severity of Subjective Cognitive Complaints and Worries in Older Adults Are Associated With Cerebral Amyloid-β Load. Frontiers in Aging Neuroscience, 13, 675583. 10.3389/fnagi.2021.675583 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Song Y, Wu F, Sharma S, Clendenen TV, India-Aldana S, Afanasyeva Y, Gu Y, Koenig KL, Zeleniuch-Jacquotte A, & Chen Y. (2023). Mid-life adherence to the Dietary Approaches to Stop Hypertension (DASH) diet and late‐life subjective cognitive complaints in women. Alzheimer’s & Dementia, alz. 13468. 10.1002/alz.13468 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Stringa N, van Schoor NM, Milaneschi Y, Ikram MA, Del Panta V, Koolhaas CM, Voortman T, Bandinelli S, Wolters FJ, & Huisman M. (2020). Physical Activity as Moderator of the Association Between APOE and Cognitive Decline in Older Adults: Results from Three Longitudinal Cohort Studies. The Journals of Gerontology: Series A, 75(10), 1880–1886. 10.1093/gerona/glaa054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Tan ZS, Spartano NL, Beiser AS, DeCarli C, Auerbach SH, Vasan RS, & Seshadri S. (2016). Physical Activity, Brain Volume, and Dementia Risk: The Framingham Study. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, glw130. 10.1093/gerona/glw130 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Tang T, Austin PC, Lawson KA, Finelli A, & Saarela O. (2020). Constructing inverse probability weights for institutional comparisons in healthcare. Statistics in Medicine, 39(23), 3156–3172. 10.1002/sim.8657 [DOI] [PubMed] [Google Scholar]
  54. Toniolo PG, Levitz M, Zeleniuch-Jacquotte A, Banerjee S, Koenig KL, Shore RE, Strax P, & Pasternack BS. (1995). A Prospective Study of Endogenous Estrogens and Breast Cancer in Postmenopausal Women. JNCI Journal of the National Cancer Institute, 87(3), 190–197. 10.1093/jnci/87.3.190 [DOI] [PubMed] [Google Scholar]
  55. Toniolo PG, Pasternack BS, Shore RE, Sonnenschein E, Koenig KL, Rosenberg C, Strax P, & Strax S. (1991). Endogenous hormones and breast cancer: A prospective cohort study. Breast Cancer Research and Treatment, 18(S1), S23–S26. 10.1007/BF02633522 [DOI] [PubMed] [Google Scholar]
  56. Walker E, Wu F, Rundle AG, Hua S, Quinn JW, Neckerman KM, Afanasyeva Y, Arslan AA, Koenig KL, Zeleniuch-Jacquotte A, & Chen Y. (2026). The association between midlife neighbourhood walkability and Alzheimer’s disease in women: A prospective nested case–control study. Age and Ageing, 55(3), afag054. 10.1093/ageing/afag054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Weuve J. (2004). Physical Activity, Including Walking, and Cognitive Function in Older Women. JAMA, 292(12), 1454. 10.1001/jama.292.12.1454 [DOI] [PubMed] [Google Scholar]
  58. World Health Organization. (2022). Global status report on physical activity 2022. World Health Organization. https://apps.who.int/iris/handle/10665/363607 [Google Scholar]
  59. Wu F, Davey S, Clendenen TV, Koenig KL, Afanasyeva Y, Zhou B, Bedi S, Li H, Zeleniuch-Jacquotte A, & Chen Y. (2022). Gut Microbiota and Subjective Memory Complaints in Older Women. Journal of Alzheimer’s Disease, 88(1), 251–262. 10.3233/JAD-220011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zhang K, Wu F, Clendenen TV, Afanasyeva Y, Gu Y, Koenig KL, Zeleniuch-Jacquotte A, & Chen Y. (2025). Association between midlife intake of ultra-processed foods and later-life subjective cognitive complaints: Evidence from the new York University women’s health study. Preventive Medicine, 200, 108390. 10.1016/j.ypmed.2025.108390 [DOI] [PubMed] [Google Scholar]
  61. Zhong W-F, Wang X-M, Chen H, Song W-Q, Gao J, Chen P-L, Shen Q-Q, You F-F, Li C, Lv Y-B, Li Z-H, Shi X-M, & Mao C. (2025). Combined effects of physical activity, cognitive activity, and dietary patterns on cognitive impairment in older adults with consideration of APOE genotype: A cohort study. Alzheimer’s Research & Therapy, 17(1), 158. 10.1186/s13195-025-01806-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zsuffa JA, Katz S, Koszovacz V, Berente DB, Kamondi A, Csukly G, Mangialasche F, Rocha ASL, Kivipelto M, & Horvath AA. (2024). Lifestyle and behavioural changes in older adults during the Covid-19 pandemic are associated with subjective cognitive complaints. Scientific Reports, 14(1), 2502. 10.1038/s41598-024-52856-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material
NIHMS2210952-supplement-Supplementary_Material.docx (26.8KB, docx)

RESOURCES