
==== Front
medRxiv
MEDRXIV
medRxiv
Cold Spring Harbor Laboratory

39148839
10.1101/2024.08.05.24311482
preprint
2
Article
The Interaction Effects of Sex, Age, APOE and Common Health Risk Factors on Human Brain Functions
http://orcid.org/0000-0001-6142-3865
Li Tengfei PhD 12†
Chen Jie MS 3†
Zhao Bingxin PhD 4
Garden Gwenn A. MD, PhD 5
Giovanello Kelly S. PhD 6
Wu Guorong PhD 6789*
Zhu Hongtu PhD 13710*
1 Biomedical Research Imaging Center, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
2 Department of Radiology, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
3 Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
4 Department of Statistics and Data Science, the Wharton School, University of Pennsylvania, Philadelphia, PA, USA
5 Department of Neurology, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
6 Department of Psychiatry, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
7 Department of Statistics and Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
8 UNC Neuroscience Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
9 Carolina Insititute for Developmental Disabilities, Chapel Hill, NC, USA
10 Departments of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
† These authors contributed equally

Author Contributions: Concept and design: Zhu, Wu and Li; Acquisition, analysis, or interpretation of data: Li and Chen; Drafting of the manuscript: Li and Chen; Critical revision of the manuscript for important intellectual content: Li, Chen, Wu, Zhu, Zhao, Giovanello and Garden; Statistical analysis: Chen and Li; Obtained funding: Wu and Zhu

* Corresponding authors: Hongtu Zhu (primary corresponding author), Professor, Department of Biostatistics, 3105C McGavran_Greenberg Hall, CB# 7420, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, htzhu@email.unc.edu, Guorong Wu (co-corresponding author), Associate Professor, Department of Psychiatry, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, guorong_wu@med.unc.edu
02 9 2024
2024.08.05.24311482https://creativecommons.org/licenses/by/4.0/ This work is licensed under a Creative Commons Attribution 4.0 International License, which allows reusers to distribute, remix, adapt, and build upon the material in any medium or format, so long as attribution is given to the creator. The license allows for commercial use.
nihpp-2024.08.05.24311482.pdf
Importance:

Nonlinear changes in brain function during aging are shaped by a complex interplay of factors, including sex, age, genetics, and modifiable health risk factors. However, the combined effects and underlying mechanisms of these factors on brain functional connectivity remain poorly understood.

Objective:

To comprehensively investigate the combined associations of sex, age, APOE genotypes, and ten common modifiable health risk factors with brain functional connectivities during aging.

Design, Setting, and Participants:

This analysis used data from 36,630 UK Biobank participants, aged 44–81, who were assessed for sex, age, APOE genotypes, 10 health risk factors, and brain functional connectivities through resting-state functional magnetic resonance imaging.

Main Outcomes and Measures:

Brain functional connectivities were evaluated through within- and between-network functional connectivities and connectivity strength. Associations between risk factors and brain functional connectivities, including their interaction effects, were analyzed.

Results:

Hypertension, BMI, and education were the top three influential factors. Sex-specific effects were also observed in interactions involving APOE4 gene, smoking, alcohol consumption, diabetes, BMI, and education. Notably, a negative sex-excessive alcohol interaction showed a stronger negative effect on functional connectivities in males, particularly between the dorsal attention network and the language network, while moderate alcohol consumption appeared to have protective effects. A significant negative interaction between sex and APOE4 revealed a greater reduction in functional connectivity between the cingulo-opercular network and the posterior multimodal network in male APOE4 carriers. Additional findings included a negative age-BMI interaction between the visual and dorsal attention networks, and a positive age-hypertension interaction between the frontoparietal and default mode networks.

Conclusions and Relevance:

The findings highlight significant sex disparities in the associations between age, the APOE-ε4 gene, modifiable health risk factors, and brain functional connectivity, emphasizing the necessity of jointly considering these factors to gain a deeper understanding of the complex processes underlying brain aging.

U.S. National Institute On Aging (NIA) of the National Institutes of Health (NIH)RF1AG082938 U01AG079847 NIHNS110791 MH116527 AR082684 UK Biobank resource22783 UK Biobank
==== Body
pmcINTRODUCTION

Brain structural and functional aging exhibits substantial variability, influenced by factors such as sex, age, genetics, and modifiable health risk factors (MHRFs) including socioeconomic status (SES), lifestyle, and cardiovascular risk factors (CVRFs) 1,2. Our previous research examined the impact of MHRFs and their interactions on brain structure during aging 3. Building on this, it is essential to explore how these factors collectively affect brain functionality.

Age, sex, and Apolipoprotein E (APOE) genotypes are well-studied non-modifiable risk factors affecting brain function and cognition during aging4. Cognitive domains follow different trajectories, with vocabulary remaining stable while memory, reasoning, and processing speed decline 5–7. These cognitive changes can be partially explained by aging-related alterations in brain networks, including reduced connectivity within and between networks, decreased default mode network (DMN) connectivity, and network reorganization8–16. APOE-ε4 (APOE4), a key genetic risk factor for Alzheimer’s disease (AD), is associated with altered connectivity in memory and cognitive networks, accelerated age-related connectivity loss, and sometimes increased hyperconnectivity 17–21. The effects of APOE4 differ by sex 22 and age 23, with a stronger impact on attention in women and memory and executive functions in men 24. However, most studies emphasize the vulnerability of female APOE4 carriers, with limited data on functional atrophy in males 3,22,25. The joint influence of age, sex, APOE, and other MHRFs on brain function also remains unclear.

CVRFs, lifestyle factors, and SES are extensively studied in relation to brain and cardiovascular health. Key CVRFs identified by the Framingham Heart Study—such as hypertension, smoking, cholesterol, diabetes, obesity, and family history of heart disease—can impair brain health and contribute to aging and neurodegenerative diseases 26–30. The American Heart Association’s Life’s Essential 8 highlights crucial lifestyle measures like diet, physical activity, and sleep for maintaining cardiovascular and brain health 31. Recent studies show a strong link between cardiovascular and brain aging, with cardiovascular dysfunction from risk factors potentially impairing brain health 32–34. The 2020 Lancet Commission report suggests that modifying 12 major dementia risk factors could prevent or delay up to 40% of dementia cases 35.

This study aims to perform a comprehensive association analysis between brain functional connectivity (FC) and various risk factors, including APOE genotype, age, sex, and ten MHRFs (Figure 1). These MHRFs encompass six adverse factors—hypertension 36–40, diabetes 41–43, smoking 44–46, obesity 47–49, excessive alcohol consumption 50,51, and social deprivation 52—and four beneficial ones—education 53–56, physical activity 57, healthy diet 58 and sleep 59 that may affect brain FCs. Using resting-state functional magnetic resonance imaging (rsfMRI) data from 36,630 UK Biobank (UKB) subjects, this study seeks to uncover new insights into the dynamics of aging and preclinical dementia risks, particularly focusing on sex disparities.

METHODS

Study Population

The UKB dataset comprises around 500,000 participants, including over 40,000 who underwent MRI scans. We analyzed data from 39,354 subjects with both T1-weighted and rsfMRI scans, including 36,339 with British ancestry and 3,015 with non-British ancestry. The primary analyses focused on 36,630 unrelated individuals with complete brain imaging, genetic, and MHRF data. Key findings were based on 33,824 white British subjects, with subsequent validation in 2,806 non-British subjects.

Imaging Data Processing

Imaging data acquisition and preprocessing followed protocols outlined in the UKB Brain Imaging Documentation (https://biobank.ctsu.ox.ac.uk/crystal/crystal/docs/brain_mri.pdf). RsfMRI data underwent motion correction, intensity normalization, and ICA-based artifact removal. High-quality T1-weighted images were preprocessed and co-registered rsfMRI data to MNI standard space. The HCP-MMP 60 atlas was employed to generate 64,620 functional connectivity (FC) traits per subject, which were then classified into twelve resting-state networks (Ji-12 network) 61, namely, the somatomotor (SMN), auditory (AN), visual1 (Vi1), visual2 (Vi2), dorsal attention (DAN), DMN, frontoparietal (FPN), language (LAN), cingulo-opercular (CON), posterior multimodal (PMN), ventral multimodal (VMN), and orbito-Affective (OAN) networks. For each rsfMRI scan, the mean time series from each of 360 ROIs were extracted and the correlation between each pair of regional time series was transformed from Pearson correlations to z-statistics using Fisher transformation. For each pair of the 12 networks we calculated the between- and within-network functional connectivity (NFC; or mean FC) and edge strength (NES; or mean absolute value of FC) measures. We also generated 100 × 100 FC matrices based on Schaefer 100 parcellation atlas and network-level traits for the Yeo-7 and Yeo-17 atlases 62. Detailed imaging protocol and processing steps were provided in Supplement 1.

APOE Genotyping and covariates

APOE haplotypes were determined using SNPs rs7412 and rs429358, defining three alleles—APOE-ε2 (APOE2), APOE-ε3, and APOE4. Details on MHRFs and other covariates are provided in Supplement 1.

Statistical Analysis

Statistical analyses were conducted using R version 4.1.0. APOE2 and APOE4 were modeled using additive models. We analyzed between- and within-network NFC and NES measures, considering APOE4 and APOE2 counts, sex, age, and MHRFs as covariates of interest. To investigate interactions between age or APOE genotype and MHRFs on FC measures, we included two-way interactions (age or APOE with MHRFs) as well as age-APOE-MHRF interactions. To assess sex differences, two-way interactions between sex and age, APOE genotype, and MHRFs were incorporated, along with sex-related three-way interactions (sex-age-MHRF, sex-APOE-MHRF, and sex-age-APOE). Each main and interaction effect was tested separately, controlling for age, sex, all the other risk factors, head motion, brain position, volumetric scaling, study site, phase, and living with a partner. Age-squared and sex-age-squared interactions were also controlled, except when sex, age, or sex-age interaction were terms of interest. Detailed models are available in eTable 2, Supplement 2. Type II ANOVA F-tests were used to assess the main and interaction effects 63. Our preliminary findings, based on white British subjects, were evaluated using Bonferroni-corrected significance levels of 0.05/m (details on the number of tests m, are provided in eTable 3, Supplement 2). Results were validated in non-British UKB populations by consistency of association directions. Post hoc analyses of specific ROIs within identified networks were conducted for each risk factor, with the same confounding covariates and Bonferroni correction methods (m being the number of ROI pairs in identified networks for each risk factor).

RESULTS

Participant Characteristics

Demographic information for the 36,630 UKB subjects is summarized in Table 1. Correlation plots for APOE gene and MHRFs (eFigure 1, Supplement 3) show weak correlations (r < 0.3 except for moderate and excessive alcohol consumption). Population-mean NFC and NES matrices for the Ji-12, Yeo-7, and Yeo-17 atlases (eFigures 43–45, Supplement 3) reveals anti-correlations between both the DMN and VMN with the AN, CON, DAN, PMN, SMN, Vi1, and Vi2 networks, suggesting competitive or complementary functions critical for cognitive processing. Demographic differences between populations are detailed in eTables 4 and 5 (Supplement 2). With a Bonferroni-corrected significance level of 0.0038, the non-British cohort was, on average, two years younger, with fewer ever-smokers (−3.3%) and fewer partnered individuals (−6.9%), but higher proportions of advanced education (14.1%), higher SoDep index (17.1%), excessive alcohol consumption (10.8%), and diabetes (1.5%).

Overall Findings

Findings for NFC and NES measures across the Ji-12, Yeo-7, and Yeo-17 atlases are summarized in eTable 6, Supplement 2, and depicted in eFigures 37–42, Supplement 3. Main effects are presented in eFigures 2–7, Supplement 3, while two-way and three-way interaction effects are shown in eFigures 8–19, Supplement 3. In the white British population, we identified 113, 54, and 123 associations between NES measures and MHRFs for the Ji-12, Yeo-7, and Yeo-17 atlases, respectively, with 89.4% or more validated in non-British populations. For NFC measures, 113, 68, and 185 associations were identified, with 80–83% validated. Effect sizes were consistent across ethnicities, demonstrating generally robust associations (eFigures 31–36, Supplement 3).

Associations between Age, Sex and APOE and Brain Functions

Figure 2A shows associations between age, sex, APOE, and FC measures. Consistent with previous research 8,64, most within- and between-network NES measures decrease with age, except for increases within the FPN and between the FPN-DAN and FPN-VMN networks (eFigure 2, Supplement 3). These exceptions align with the posterior-anterior shift in aging (PASA), reflecting increased frontal activity as posterior occipital activity declines 10,12,13. Sex differences revealed that males generally exhibited higher NES measures across most networks, except for the DMN, where females had higher connectivity, consistent with prior findings 65,66 . Positive age-sex interactions showed that males experienced more pronounced increases in FPN and LN connectivity and larger reductions in visual networks with aging, while females showed stronger reductions in connectivity between the AN, DMN, VMN, and CON networks. No significant effects were observed for the APOE2 variant, but APOE4 carriers showed reduced connectivity, particularly within the Vi2, DAN, DMN, and FPN networks, consistent with earlier studies 17. A negative sex-APOE4 interaction (Figure 2C) in the CON-PMN networks (ES: −0.020, p = 5.2E-4) suggested reduced CON-PMN connectivity in male APOE4 carriers, offering new insights. Further ROI analysis revealed reduced CON-PMN connectivity in male APOE4 carriers between L_p32pr and R_TPOJ2 (Figure 2D; ES: −0.023, p = 7.02E-05).

Associations between ten MHRFs and Brain Functions

Hypertension.

Negative effects of hypertension were widespread across networks, including AN, SMN, OAN, VMN, CON, LAN, and DMN and 26 between-network NES measures (Ji-12), supporting its role in accelerating aging 37,38. A positive age-hypertension interaction was found on NFC between the FPN and DMN (Figure 3C and eFigure 20, Supplement 3) networks (ES: 0.029, p = 4.9E-6). Further post-hoc analyses revealed a positive age-hypertension interaction between L_p9–46v and R_9m (Figure 3D and eFigure 28, Supplement 3; ES: 0.029, p = 1.2E-5), indicating reduced anti-correlation, or decreased functional segregation in participants with hypertension during aging 39,40.

BMI.

BMI exhibited complex effects on connectivity, with negative effects on NES within the OAN and 19 between-network pairs, and positive effects within the DAN and CON networks (Figures 3A and 3B, and eFigures 21–23, Supplement 3). A negative age-BMI interaction on Vi2-DAN NES suggests that higher BMI accelerates age-related decline in FC (eFigure 49, Supplement 3); negative sex-BMI interactions were found on NES within Vi1 (ES: −0.023, p = 1.2E-4), and between Vi1 and Vi2 networks (ES: −0.020, p = 6.4E-4), indicating a more rapid decline in visual connectivity strength in males (Figures 3A and 3C). Further post-hoc analyses revealed 19 ROI-level negative associations in the within-Vi1 and Vi-Vi2 networks (Figures 3C and 3D, and eFigure 29, Supplement 3), 18 of which (94.7%) were validated in non-British populations.

Diabetes.

Although no main effects of diabetes were observed, a positive sex-diabetes interaction was found on NFC between the Vi2 and AN networks (ES: 0.021, p = 5.7E-4) (Figure 3C). Further post-hoc analyses revealed positive sex-diabetes interaction effects on FCs between L_PBelt and ROIs L_V3A, L_V7, R_V3, R_V8, and R_V3B, and between R_V8 and R_A4 (Figures 3C and 3D and eFigure 30, Supplement 3), highlighting the greater negative impact of diabetes on visual-auditory connectivity in females 41.

Smoking.

Consistent with previous literature 44–46, smoking was associated with widespread negative effects, including within FPN and VMN, between VMN and six networks, and between DAN and OAN (Figures 4A and 4B). However, a positive effect was noted between the CON and AN networks. A negative sex-smoking interaction on NFC between the VMN and CON (ES: −0.020, p = 2.3E-4) suggested greater detrimental effects from smoking in males (Figures 4A and 4C). ROI-level analyses confirmed these effects between PeEc and pOFC (Figures 4C and 4D).

Alcohol.

Excessive alcohol consumption negatively impacted NES between the DMN-OAN, DMN-Vi1, and OAN-Vi2 networks (Figures 4A and 4B). A negative sex-excessive alcohol interaction showed a stronger effect in males, particularly between the DAN and LAN networks (ES: −0.021, p = 1.9E-4; Figures 4A, 4C, and eFigure 25, Supplement 3). Conversely, a positive sex-moderate alcohol interaction was observed between the DAN-LAN and FPN-PMN networks, indicating different effects across alcohol consumption levels.

Physical activity, Sleep and Education.

Positive effects of physical activity and sleep were observed within SMN networks (Figures 4A and 4B), consistent with previous studies linking sensory/somatomotor network connectivity to sleep quality 59. Education was positively associated with NES within the OAN, VMN, and LAN networks, and across 13 network pairs (eFigure 50, Supplement 3) A negative sex-education interaction impacted NFC between CON and FPN (ES: −0.021, p = 1.7E-4), while a positive interaction enhanced NES between the PMN-DMN networks (ES: 0.023, p = 4.5E-5), suggesting higher education is associated with greater CON-FPN connectivity in females and PMN-DMN connectivity in males (eFigures 26 and 27, Supplement 3).

DISCUSSION

To our knowledge, this is the largest study to systematically examine the interactions of aging, sex, the APOE gene, and MHRFs with brain functional measures. We identified significant sex differences in brain FC associated with factors such as age, APOE4, BMI, diabetes, smoking, alcohol consumption, and education. For example, smoking, alcohol, and BMI had more pronounced negative effects in males, while diabetes had a greater impact on females in specific networks. Education was linked to positive effects on CON-FPN networks in females and PMN-DMN networks in males. Additionally, APOE4’s effects were more pronounced between the CON and PMN networks in males. These findings reinforce current understandings of sex-specific effects of MHRFs and highlight the need for further explorations.

The sex-age interactions revealed that males experienced greater increases in NES within the FPN and between the DMN-LN and DAN-LN networks, along with larger reductions in the visual network. In contrast, females exhibited greater reductions in connectivity between the AN, DMN, VMN, and CON networks, suggesting distinct aging patterns: males undergoing more extensive reorganization (e.g., PASA) 10,12,13 and females demonstrating higher vulnerability to neurodegeneration 67. The negative sex-APOE4 interaction suggests that male carriers are particularly vulnerable, with more pronounced reductions in connectivity between the L_p32pr-R_TPOJ2 regions and between the CON and PMN networks—areas critical for memory and executive function. While much of the literature focuses on the cognitive vulnerability of female APOE4 carriers 3,22,25, our findings suggest that APOE4 may have a larger impact on the decline of CON-PMN brain networks related to memory and executive functions in men 24. The sex-diabetes interaction suggests that diabetes affects brain connectivity differently by sex, with females showing a greater change in visual-auditory connectivity 41. A negative sex-smoking interaction was observed between the VMN and CON networks, indicating more severe smoking-related impairments in males. Similarly, the negative sex-excessive alcohol interaction showed a larger detrimental effect between the DAN and LAN networks in males, likely due to alcohol misuse leading to lower cortical volume, reduced white matter and hippocampal volume, and greater changes in brain function and behavior in men 68–70. The sex-BMI interaction highlights a larger impact of obesity on brain function in males, aligning with previous research that suggests men experience detrimental changes in brain connectivity starting from the overweight category, while women typically show declines only in the obese range, possibly due to obesity-induced chronic white matter damage in males 71.

Our findings on excessive alcohol consumption corroborate previous research, indicating that reduced connectivity in the precuneus, postcentral gyrus, insula, visual cortex, and left executive control network are key areas of rsfMRI NC reduction 50,51. Meanwhile, our study highlights a sex-specific dichotomy between excessive and moderate alcohol consumption on brain connectivity. In males, excessive alcohol had more detrimental effects between the DAN and LAN, while moderate consumption showed a protective effect on connectivity between the DAN and LAN, and FPN and PMN networks, suggesting a U-shaped nonlinear patterns between alcohol dose and brain health 72.

We observed that higher education positively influences connectivity within and between networks including the OAN, VMN, LAN, and FPN, potentially enhancing cognitive reserve 53,54. Sex-education interactions revealed that with higher education, females showed increased connectivity in the CON-FPN network, while males benefited more in the PMN-DMN network. This suggests that education may enhance neural pathways aligned with the cognitive needs of each sex, indicating the potential for tailored educational programs to optimize brain health and cognitive function.

Elevated blood pressure, a common cardiovascular risk factor, is linked to cognitive decline in later life. Hypertension broadly impairs connectivity within and between attentional, SMN, and DMN networks, likely due to its impact on neural inefficiencies 73. We observed reduced anti-correlation between the dorsolateral (L_p9–46v) and medial (R_9m) prefrontal cortices, suggesting decreased functional segregation. A recent study 39 demonstrated that higher blood pressure causally reduces brain functional segregation and worsening cognition in the aging population through observational and Mendelian randomization analyses.

Strength and Limitations

Leveraging large-sample fMRI data from the UKB, we investigated a broad range of modifiable and non-modifiable risk factors, exploring their joint, conditional, and interaction effects on brain functions. These findings were validated across multiple atlases and ethnic populations, ensuring robustness against racial differences, atlas choices, outlier sensitivities, and sample size limitations. Our main analyses were based on parcellation-based full correlations. Although FMRIB’s ICA-based X-noiseifier (FIX) has been applied to the UKB dataset to remove scanner artifacts and motion effects, full correlation measures can be sensitive to remaining global artifacts 74, while measuring partial functional connectivity between paired brain regions can reduce global artifacts and remove dependencies on other brain regions 75. Future studies will explore parcellation-based partial correlation traits.

CONCLUSION

Our study revealed sex differences in the effects of APOE4 and MHRFs on brain FC measures, with male APOE4 carriers experiencing larger declines in FC between the CON and PMN networks. Distinct aging patterns emerged, where males showed more neural reorganization, and females demonstrated greater vulnerability to neurodegeneration. Sex-specific effects of diabetes, smoking, BMI, and education underscore the importance of considering sex and demographic factors in brain health research. Additionally, we observed a sex-specific dichotomy in the impact of alcohol consumption: excessive drinking reduces, while moderate consumption increases functional connectivity strength between dorsal attention and language networks in males.

Supplementary Material

Supplement 1

Funding/Support:

This work was supported in part by the U.S. National Institute On Aging (NIA) of the National Institutes of Health (NIH) grant (RF1AG082938, U01AG079847) and NIH grants (NS110791, MH116527, AR082684). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Role of the Funder/Sponsor:

This research has been conducted using the UK Biobank resource (application number 22783), subject to a data transfer agreement. As such, the investigators within the UK Biobank contributed to the data but did not participate in analysis or writing of this report. We thank the individuals represented in the UK Biobank, studies for their participation and the research teams for their work in collecting, processing and disseminating these datasets for analysis.

Data Sharing Statement:

The individual-level data used in this study can be obtained from https://www.ukbiobank.ac.uk/.

Figure 1. The study design.

(A) Multimodal data in the UK Biobank. (B) Schematic diagram illustrating the analyzed associations between the APOE gene, demographics, modifiable health risk factors, and brain functional traits. (C) Identified brain network pairs based on the Ji-12 atlas, showing network edge strength (NES) measures associated with APOE, alcohol, smoking, BMI, physical activity, diet, sleep, education, diabates and hypertension (main effects only; no interactions considered). SMN, somatomotor network; AN, auditory network; Vi1, Vi2, visual networks 1 and 2; DAN, dorsal attention network; DMN, default mode network; FPN, frontoparietal network; LN, language network; CON, cingulo-opercular network; PMN/VMN, posterior/ventral multimodal network; OAN, orbito-affective network; SES, socioeconomic status (such as education); Demog: demographics (age and sex); EDU: education; APOE: apolipoprotein E; CVRF: cardiovascular risk factors; BMI, Body Mass Index.

Figure 2. Selected associations of APOE4 with brain functional network connectivity measures.

(A) Heatmaps show association results from the white British population using three network atlases: Ji-12, Yeo-7, and Yeo-17, for network edge strength (NES; I) and functional connectivity (NFC; II). Significant results validated in the non-British population are marked with (*); significant but unvalidated results are indicated by (+). Non-significant results but with p-values <1e-4 and <1e-3 are denoted by (.) and (..), respectively. (B) Circular plots showing NES (I) and NFC (II) associations from the three atlases. Network spatial locations shown on the right, with colored spheres representing different networks and red and blue lines for positive and negative associations. (C) Boxplots and scatterplots illustrating APOE-ε4 interaction effects on network and regional connectivity: Row 1 shows sex-APOE effect on the CON-PMN NFC (Ji-12) and age-APOE-education effects on within-DMN NES (Yeo-7) and DM3-DM4 NES (Yeo-17); Row 2 shows sex-APOE effects on connectivity between L_P32pr and R_TPOJ2, and age-APOE-education effects on connectivity between Ld32 and R47I, and between L_23d and R_47I. (D) Spatial locations of brain regions for sex-APOE4 and age-APOE4-education interactions.

Figure 3. Selected associations of hypertension, BMI, and diabetes with brain functional network connectivity measures.

(A) Heatmaps showing association results from the white British population using Ji-12, Yeo-7, and Yeo-17, for network edge strength (NES; I) and network functional connectivity (NFC; II). Significant results validated in the non-British population are marked with (*); significant but unvalidated results are indicated by (+). Non-significant results but with p-values <1e-4 and <1e-3 are denoted by (.) and (..), respectively. (B) Circular plots showing NES (I) and NFC (II) associations from the Ji-12 atlas. Colors spheres represent different networks with red and blue lines for positive and negative associations. (C) Scatter plots and boxplots illustrating interaction effects. Columns 1–3: age-hypertension effects on FPN-DMN NFC, sex-BMI effects on Vi2-Vi2 NES, and sex-diabetes effects on Vi2-AN NFC; Columns 4–6: age-hypertension effects on connectivity between L_p9–46v and R_9m, sex-BMI effects on connectivity between L_PIT and L_ProS, and sex-diabetes effects on connectivity between L_V3A and L_Pbelt. (D) Spatial locations of the identified brain regions for the sex-diabetes, age-hypertension, and sex-BMI interactions.

Figure 4. Selected associations of smoking, alcohol consumption, physical activity and sleep with brain functional network connectivity measures.

(A) Heatmaps showing association results from the white British population using Ji-12, Yeo-7, and Yeo-17, for network edge strength (NES; I) and network functional connectivity (NFC; II). Significant results validated in the non-British population are marked with (*); significant but unvalidated results are indicated by (+). Non-significant results but with p-values <1e-4 and <1e-3 are denoted by (.) and (..), respectively. (B) Circular plots showing NES (I) and NFC (II) associations from the three atlases. Colors spheres represent different networks with red and blue lines for positive and negative associations. (C) Scatter plots and boxplots illustrating interaction effects. Row 1, columns 1–3: the sex-smoking interaction effect on the OAN-VMN between-network NFC, sex-excessive alcohol interaction effect on the DAN-LAN between-network NES, and sex-excessive alcohol interaction effect on the DM1-DM4 between-network NES (Yeo-17 atlas), respectively; row 2, columns 1–3: sex-smoking effects on brain functional connectivities between brain regions L_PeEc and L_pOFC, between L_pOFC and R_PeEc, and between R_9m and R_pOFC, respectively. (D) Spatial locations of the identified brain regions for sex-smoking interactions.

Table 1. Demographic information, APOE gene, and modifiable health-related risk factors for 36,630 UK Biobank subjects.

Characteristic	Female (N=19395)	Male (N=17235)	Total (N=36630)	
Age at imaging: mean (SD) [range]	63.0 (7.39)
[45.0, 81.0]	64.3 (7.65)
[44.0, 81.0]	63.6 (7.54)
[44.0, 81.0]	
Ethnicity---British: No. (%)	17861 (92.1%)	15963 (92.6%)	33824 (92.3%)	
Ethnicity---non-British: No. (%)	1534 (7.9%)	1272 (7.4%)	2806 (7.7%)	
APOE4 = 0: No. (%)	13868 (71.5%)	12576 (73.0%)	26444 (72.2%)	
APOE4 = 1: No. (%)	5085 (26.2%)	4295 (24.9%)	9380 (25.6%)	
APOE4 = 2: No. (%)	442 (2.3%)	364 (2.1%)	806 (2.2%)	
APOE2 = 0: No. (%)	16378 (84.4%)	14619 (84.8%)	30997 (84.6%)	
APOE2 = 1: No. (%)	2904 (15.0%)	2518 (14.6%)	5422 (14.8%)	
APOE2 = 2: No. (%)	113 (0.6%)	98 (0.6%)	211 (0.6%)	
Marriage---living with a partner: No. (%)	13619 (70.2%)	13913 (80.7%)	27532 (75.2%)	
Marriage---not with a partner: No. (%)	1235 (6.4%)	520 (3.0%)	1755 (4.8%)	
Education---college degree/above: No. (%)	8626 (44.5%)	8293 (48.1%)	16919 (46.2%)	
Education---no college/below: No. (%)	9433 (48.6%)	7737 (44.9%)	17170 (46.9%)	
SoDep Index---above the median: No. (%)	8275 (42.7%)	7092 (41.1%)	15367 (42.0%)	
SoDep Index---below the median: No. (%)	11099 (57.2%)	10130 (58.8%)	21229 (58.0%)	
Smoking status---ever smoked: No. (%)	12459 (64.2%)	9867 (57.2%)	22326 (61.0%)	
Smoking status---never smoked: No. (%)	6936 (35.8%)	7368 (42.8%)	14304 (39.1%)	
Regular physical activity---yes: No. (%)	14199 (73.2%)	13267 (77.0%)	27466 (75.0%)	
Regular physical activity---no: No. (%)	5040 (26.0%)	3893 (22.6%)	8933 (24.4%)	
Health diet---yes: No. (%)	10517 (54.2%)	7371 (42.8%)	17888 (48.8%)	
Health diet---no: No. (%)	8870 (45.7%)	9854 (57.2%)	18724 (51.1%)	
Alcohol consumption---excessive: No. (%)	2742 (14.1%)	1284 (7.5%)	4026 (11.0%)	
Alcohol consumption---moderate: No. (%)	11115 (57.3%)	11096 (64.4%)	22211 (60.6%)	
Alcohol consumption---not current: No. (%)	5531 (28.5%)	4848 (28.1%)	10379 (28.3%)	
Sleeping---between 6-8 hours: No. (%)	8193 (42.2%)	7759 (45.0%)	15952 (43.5%)	
Sleeping---outside 6-8 hours: No. (%)	11145 (57.5%)	9455 (54.9%)	20600 (56.2%)	
Hypertension---yes: No. (%)	3060 (15.8%)	4259 (24.7%)	7319 (20.0%)	
Hypertension---no: No. (%)	16335 (84.2%)	12976 (75.3%)	29311 (80.0%)	
Diabetes---yes: No. (%)	573 (3.0%)	975 (5.7%)	1548 (4.2%)	
Diabetes---no: No. (%)	18822 (97.0%)	16260 (94.3%)	35082 (95.8%)	
BMI: mean (SD) [range]	26.0 (4.53)
[14.7,56.6]	27.1 (3.73)
[16.7,56.0]	26.5 (4.20)
[14.7,56.6]	
No., sample size; SD, standard deviation; SoDep, social deprivation.

Conflict of Interest Disclosures: Authors have no conflicts of interest to disclose
==== Refs
REFERENCES

1. Bethlehem R a. I , Seidlitz J , White SR , Brain charts for the human lifespan. Nature. 2022;604 (7906 ):525–533. doi:10.1038/s41586-022-04554-y 35388223
2. Sun L , Zhao T , Liang X , Functional connectome through the human life span. Published online June 10, 2024:2023.09.12.557193. doi:10.1101/2023.09.12.557193
3. Chen J , Li T , Zhao B , The interaction effects of age, APOE and common environmental risk factors on human brain structure. Cereb Cortex. 2024;34 (1 ):bhad472. doi:10.1093/cercor/bhad472 38112569
4. Loeffler DA . Modifiable, Non-Modifiable, and Clinical Factors Associated with Progression of Alzheimer’s Disease. J Alzheimers Dis JAD. 2021;80 (1 ):1–27. doi:10.3233/JAD-201182 33459643
5. Harada CN , Natelson Love MC , Triebel K . Normal Cognitive Aging. Clin Geriatr Med. 2013;29 (4 ):737–752. doi:10.1016/j.cger.2013.07.002 24094294
6. Kavé G. Vocabulary changes in adulthood: Main findings and methodological considerations. Int J Lang Commun Disord. 2024;59 (1 ):58–67. doi:10.1111/1460-6984.12820 36415955
7. Brito DVC , Esteves F , Rajado AT , Assessing cognitive decline in the aging brain: lessons from rodent and human studies. Npj Aging. 2023;9 (1 ):1–11. doi:10.1038/s41514-023-00120-6 36697402
8. Archer JA , Lee A , Qiu A , Chen SHA . A Comprehensive Analysis of Connectivity and Aging Over the Adult Life Span. Brain Connect. 2016;6 (2 ):169–185. doi:10.1089/brain.2015.0345 26652914
9. Sala-Llonch R , Bartrés-Faz D , Junqué C . Reorganization of brain networks in aging: a review of functional connectivity studies. Front Psychol. 2015;6 :663. doi:10.3389/fpsyg.2015.00663 26052298
10. Davis SW , Dennis NA , Daselaar SM , Fleck MS , Cabeza R . Que PASA? The posterior-anterior shift in aging. Cereb Cortex N Y N 1991. 2008;18 (5 ):1201–1209. doi:10.1093/cercor/bhm155
11. Belden A , Quinci MA , Geddes M , Donovan NJ , Hanser SB , Loui P . Functional Organization of Auditory and Reward Systems in Aging. J Cogn Neurosci. 2023;35 (10 ):1570–1592. doi:10.1162/jocn_a_02028 37432735
12. Grady C , Sarraf S , Saverino C , Campbell K . Age differences in the functional interactions among the default, frontoparietal control, and dorsal attention networks. Neurobiol Aging. 2016;41 :159–172. doi:10.1016/j.neurobiolaging.2016.02.020 27103529
13. Matthäus F , Schmidt JP , Banerjee A , Schulze TG , Demirakca T , Diener C . Effects of age on the structure of functional connectivity networks during episodic and working memory demand. Brain Connect. 2012;2 (3 ):113–124. doi:10.1089/brain.2012.0077 22698449
14. Sambataro F , Murty VP , Callicott JH , Age-related alterations in default mode network: impact on working memory performance. Neurobiol Aging. 2010;31 (5 ):839–852. doi:10.1016/j.neurobiolaging.2008.05.022 18674847
15. Raykov PP , Knights E , Cam-CAN, Henson RN . Does functional system segregation mediate the effects of lifestyle on cognition in older adults? Neurobiol Aging. 2024;134 :126–134. doi:10.1016/j.neurobiolaging.2023.11.009 38070445
16. Pedersen R , Geerligs L , Andersson M , When functional blurring becomes deleterious: Reduced system segregation is associated with less white matter integrity and cognitive decline in aging. NeuroImage. 2021;242 :118449. doi:10.1016/j.neuroimage.2021.118449 34358662
17. Turney IC , Chesebro AG , Rentería MA , APOE ε4 and resting-state functional connectivity in racially/ethnically diverse older adults. Alzheimers Dement Diagn Assess Dis Monit. 2020;12 (1 ):e12094. doi:10.1002/dad2.12094
18. Trachtenberg AJ , Filippini N , Mackay CE . The effects of APOE-ε4 on the BOLD response. Neurobiol Aging. 2012;33 (2 ):323–334. doi:10.1016/j.neurobiolaging.2010.03.009 20409610
19. Machulda MM , Jones DT , Vemuri P , Effect of APOE ε4 status on intrinsic network connectivity in cognitively normal elderly subjects. Arch Neurol. 2011;68 (9 ):1131–1136. doi:10.1001/archneurol.2011.108 21555604
20. Brown JA , Terashima KH , Burggren AC , Brain network local interconnectivity loss in aging APOE-4 allele carriers. Proc Natl Acad Sci. 2011;108 (51 ):20760–20765. doi:10.1073/pnas.1109038108 22106308
21. Contreras JA , Fujisaki K , Ortega NE , Functional hyperconnectivity is associated with higher levels of sPDGFRβ, a marker of blood brain barrier breakdown, in older adults. Alzheimers Dement. 2023;19 (S3 ):e060710. doi:10.1002/alz.060710
22. Damoiseaux JS , Seeley WW , Zhou J , Gender Modulates the APOE ε4 Effect in Healthy Older Adults: Convergent Evidence from Functional Brain Connectivity and Spinal Fluid Tau Levels. J Neurosci. 2012;32 (24 ):8254–8262. doi:10.1523/JNEUROSCI.0305-12.2012 22699906
23. Bretsky P , Guralnik JM , Launer L , Albert M , Seeman TE . The role of APOE-ε4 in longitudinal cognitive decline. Neurology. 2003;60 (7 ):1077–1081. doi:10.1212/01.WNL.0000055875.26908.24 12682309
24. Fleisher AS , Chen K , Liu X , Apolipoprotein E ε4 and age effects on florbetapir positron emission tomography in healthy aging and Alzheimer disease. Neurobiol Aging. 2013;34 (1 ):1–12. doi:10.1016/j.neurobiolaging.2012.04.017 22633529
25. Paranjpe MD , Wang JK , Zhou Y . Sex, ApoE4 and Alzheimer’s disease: rethinking drug discovery in the era of precision medicine. Neural Regen Res. 2021;16 (9 ):1764–1765. doi:10.4103/1673-5374.306070 33510067
26. Gordon T , Kannel WB . Multiple risk functions for predicting coronary heart disease: the concept, accuracy, and application. Am Heart J. 1982;103 (6 ):1031–1039. doi:10.1016/0002-8703(82)90567-1 7044082
27. Gordon T , Castelli WP , Hjortland MC , Kannel WB , Dawber TR . Diabetes, blood lipids, and the role of obesity in coronary heart disease risk for women. The Framingham study. Ann Intern Med. 1977;87 (4 ):393–397. doi:10.7326/0003-4819-87-4-393 199096
28. Anderson KM , Odell PM , Wilson PW , Kannel WB . Cardiovascular disease risk profiles. Am Heart J. 1991;121 (1 Pt 2 ):293–298. doi:10.1016/0002-8703(91)90861-b 1985385
29. Wilson PWF , D’Agostino RB , Levy D , Belanger AM , Silbershatz H , Kannel WB . Prediction of Coronary Heart Disease Using Risk Factor Categories. Circulation. 1998;97 (18 ):1837–1847. doi:10.1161/01.CIR.97.18.1837 9603539
30. Kannel WB , Dawber TR , Kagan A , Revotskie N , Stokes J . Factors of risk in the development of coronary heart disease--six year follow-up experience. The Framingham Study. Ann Intern Med. 1961;55 :33–50. doi:10.7326/0003-4819-55-1-33 13751193
31. Lloyd-Jones DM , Allen NB , Anderson CAM , Life’s Essential 8: Updating and Enhancing the American Heart Association’s Construct of Cardiovascular Health: A Presidential Advisory From the American Heart Association. Circulation. 2022;146 (5 ):e18–e43. doi:10.1161/CIR.0000000000001078 35766027
32. Zhao B , Li T , Fan Z , Heart-brain connections: Phenotypic and genetic insights from magnetic resonance images. Science. 2023;380 (6648 ):abn6598. doi:10.1126/science.abn6598 37262162
33. van der Velpen IF , Yancy CW , Sorond FA , Sabayan B . Impaired Cardiac Function and Cognitive Brain Aging. Can J Cardiol. 2017;33 (12 ):1587–1596. doi:10.1016/j.cjca.2017.07.008 28966021
34. Vanherle L , Matuskova H , Don-Doncow N , Uhl FE , Meissner A . Improving Cerebrovascular Function to Increase Neuronal Recovery in Neurodegeneration Associated to Cardiovascular Disease. Front Cell Dev Biol. 2020;8 . doi:10.3389/fcell.2020.00053
35. Livingston G , Huntley J , Sommerlad A , Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet Lond Engl. 2020;396 (10248 ):413–446. doi:10.1016/S0140-6736(20)30367-6
36. Carnevale L , Maffei A , Landolfi A , Grillea G , Carnevale D , Lembo G . Brain Functional Magnetic Resonance Imaging Highlights Altered Connections and Functional Networks in Patients With Hypertension. Hypertension. 2020;76 (5 ):1480–1490. doi:10.1161/HYPERTENSIONAHA.120.15296 32951470
37. Hung TH , Chen VCH , Chuang YC , Investigating the effect of hypertension on vascular cognitive impairment by using the resting-state functional connectome. Sci Rep. 2024;14 (1 ):4580. doi:10.1038/s41598-024-54996-9 38403657
38. Tzourio C , Laurent S , Debette S . Is Hypertension Associated With an Accelerated Aging of the Brain? Hypertension. 2014;63 (5 ):894–903. doi:10.1161/HYPERTENSIONAHA.113.00147 24566080
39. Neitzel J , Malik R , Muetzel R , Genetic variants link lower segregation of brain networks to higher blood pressure and worse cognition within the general aging population. Published online August 13, 2021:2021.08.12.21261975. doi:10.1101/2021.08.12.21261975
40. Tripathi V , Batta I , Zamani A , Default mode network anti-correlation as a transdiagnostic biomarker of cognitive function. Published online April 14, 2024. doi:10.31234/osf.io/uhs3c
41. Lopez-Vilaret KM , Fernandez-Alvarez M , Shokri-Kojori E , Tomasi D , Cantero JL , Atienza M . Pre-diabetes is associated with altered functional connectivity density in cortical regions of the default-mode network. Front Aging Neurosci. 2022;14 . doi:10.3389/fnagi.2022.1034355
42. Cui Y , Li SF , Gu H , Disrupted Brain Connectivity Patterns in Patients with Type 2 Diabetes. Am J Neuroradiol. 2016;37 (11 ):2115–2122. doi:10.3174/ajnr.A4858 27365332
43. Guo X , Wang S , Chen YC , Aberrant Brain Functional Connectivity Strength and Effective Connectivity in Patients with Type 2 Diabetes Mellitus. J Diabetes Res. 2021;2021 (1 ):5171618. doi:10.1155/2021/5171618 34877358
44. Cheng W , Rolls ET , Robbins TW , Decreased brain connectivity in smoking contrasts with increased connectivity in drinking. eLife. 8 :e40765. doi:10.7554/eLife.40765 30616717
45. Yip SW , Lichenstein SD , Garrison K , Effects of Smoking Status and State on Intrinsic Connectivity. Biol Psychiatry Cogn Neurosci Neuroimaging. 2022;7 (9 ):895–904. doi:10.1016/j.bpsc.2021.02.004 33618016
46. Weiland BJ , Sabbineni A , Calhoun VD , Welsh RC , Hutchison KE . Reduced executive and default network functional connectivity in cigarette smokers. Hum Brain Mapp. 2015;36 (3 ):872–882. doi:10.1002/hbm.22672 25346448
47. Lee H , Kwon J , eun Lee J , B yong Park , Park H . Disrupted stepwise functional brain organization in overweight individuals. Commun Biol. 2022;5 (1 ):1–9. doi:10.1038/s42003-021-02957-7 34987157
48. Syan SK , McIntyre-Wood C , Minuzzi L , Hall G , McCabe RE , MacKillop J . Dysregulated resting state functional connectivity and obesity: A systematic review. Neurosci Biobehav Rev. 2021;131 :270–292. doi:10.1016/j.neubiorev.2021.08.019 34425125
49. Geha P , Cecchi G , Todd Constable R , Abdallah C , Small DM . Reorganization of brain connectivity in obesity. Hum Brain Mapp. 2016;38 (3 ):1403–1420. doi:10.1002/hbm.23462 27859973
50. Vergara VM , Liu J , Claus ED , Hutchison K , Calhoun V . Alterations of resting state functional network connectivity in the brain of nicotine and alcohol users. NeuroImage. 2017;151 :45–54. doi:10.1016/j.neuroimage.2016.11.012 27864080
51. Weiland BJ , Sabbineni A , Calhoun VD , Reduced Left Executive Network Functional Connectivity Is Associated With Alcohol Use Disorders. Alcohol Clin Exp Res. 2014;38 (9 ):2445–2453. doi:10.1111/acer.12505 25257293
52. Xiong Y , Hong H , Liu C , Zhang YQ . Social isolation and the brain: effects and mechanisms. Mol Psychiatry. 2023;28 (1 ):191–201. doi:10.1038/s41380-022-01835-w 36434053
53. Arenaza-Urquijo EM , Landeau B , La Joie R , Relationships between years of education and gray matter volume, metabolism and functional connectivity in healthy elders. NeuroImage. 2013;83 :450–457. doi:10.1016/j.neuroimage.2013.06.053 23796547
54. Stern Y. Cognitive reserve. Neuropsychologia. 2009;47 (10 ):2015–2028. doi:10.1016/j.neuropsychologia.2009.03.004 19467352
55. Shen X , Cox SR , Adams MJ , Resting-State Connectivity and Its Association With Cognitive Performance, Educational Attainment, and Household Income in the UK Biobank. Biol Psychiatry Cogn Neurosci Neuroimaging. 2018;3 (10 ):878–886. doi:10.1016/j.bpsc.2018.06.007 30093342
56. Chan MY , Han L , Carreno CA , Long-term prognosis and educational determinants of brain network decline in older adult individuals. Nat Aging. 2021;1 (11 ):1053–1067. doi:10.1038/s43587-021-00125-4 35382259
57. Bray NW , Pieruccini-Faria F , Bartha R , Doherty TJ , Nagamatsu LS , Montero-Odasso M . The effect of physical exercise on functional brain network connectivity in older adults with and without cognitive impairment. A systematic review. Mech Ageing Dev. 2021;196 :111493. doi:10.1016/j.mad.2021.111493 33887281
58. Gaynor AM , Varangis E , Song S , Diet moderates the effect of resting state functional connectivity on cognitive function. Sci Rep. 2022;12 (1 ):16080. doi:10.1038/s41598-022-20047-4 36167961
59. Bai Y , Tan J , Liu X , Cui X , Li D , Yin H . Resting-state functional connectivity of the sensory/somatomotor network associated with sleep quality: evidence from 202 young male samples. Brain Imaging Behav. 2022;16 (4 ):1832–1841. doi:10.1007/s11682-022-00654-5 35381969
60. Glasser MF , Coalson TS , Robinson EC , A multi-modal parcellation of human cerebral cortex. Nature. 2016;536 (7615 ):171–178. doi:10.1038/nature18933 27437579
61. Ji JL , Spronk M , Kulkarni K , Repovš G , Anticevic A , Cole MW . Mapping the human brain’s cortical-subcortical functional network organization. NeuroImage. 2019;185 :35–57. doi:10.1016/j.neuroimage.2018.10.006 30291974
62. Thomas Yeo BT , Krienen FM , Sepulcre J , The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J Neurophysiol. 2011;106 (3 ):1125–1165. doi:10.1152/jn.00338.2011 21653723
63. Langsrud Ø. ANOVA for unbalanced data: Use Type II instead of Type III sums of squares. Stat Comput. 2003;13 (2 ):163–167. doi:10.1023/A:1023260610025
64. Ferreira LK , Busatto GF . Resting-state functional connectivity in normal brain aging. Neurosci Biobehav Rev. 2013;37 (3 ):384–400. doi:10.1016/j.neubiorev.2013.01.017 23333262
65. Biswal BB , Mennes M , Zuo XN , Toward discovery science of human brain function. Proc Natl Acad Sci U S A. 2010;107 (10 ):4734–4739. doi:10.1073/pnas.0911855107 20176931
66. Ritchie SJ , Cox SR , Shen X , Sex Differences in the Adult Human Brain: Evidence from 5216 UK Biobank Participants. Cereb Cortex. 2018;28 (8 ):2959–2975. doi:10.1093/cercor/bhy109 29771288
67. Zhang C , Cahill ND , Arbabshirani MR , White T , Baum SA , Michael AM . Sex and Age Effects of Functional Connectivity in Early Adulthood. Brain Connect. 2016;6 (9 ):700–713. doi:10.1089/brain.2016.0429 27527561
68. Shokri-Kojori E , Tomasi D , Wiers CE , Wang GJ , Volkow ND . Alcohol affects brain functional connectivity and its coupling with behavior: greater effects in male heavy drinkers. Mol Psychiatry. 2017;22 (8 ):1185–1195. doi:10.1038/mp.2016.25 27021821
69. Ceylan-Isik AF , McBride SM , Ren J . Sex Difference in Alcoholism: Who is at a Greater Risk for Development of Alcoholic Complication? Life Sci. 2010;87 (5–6 ):133–138. doi:10.1016/j.lfs.2010.06.002 20598716
70. Flores-Bonilla A , Richardson HN . Sex Differences in the Neurobiology of Alcohol Use Disorder. Alcohol Res Curr Rev. 2020;40 (2 ):04. doi:10.35946/arcr.v40.2.04
71. Rahmani F , Wang Q , McKay NS , Sex-Specific Patterns of Body Mass Index Relationship with White Matter Connectivity. J Alzheimers Dis. 2022;86 (4 ):1831–1848. doi:10.3233/JAD-215329 35180116
72. Graff-Iversen S , Jansen MD , Hoff DA , Divergent associations of drinking frequency and binge consumption of alcohol with mortality within the same cohort. J Epidemiol Community Health. 2013;67 (4 ):350–357. doi:10.1136/jech-2012-201564 23235547
73. Mentis MJ , Salerno J , Horwitz B , Reduction of functional neuronal connectivity in long-term treated hypertension. Stroke. 1994;25 (3 ):601–607. doi:10.1161/01.str.25.3.601 8128513
74. Griffanti L , Salimi-Khorshidi G , Beckmann CF , ICA-based artefact removal and accelerated fMRI acquisition for improved resting state network imaging. NeuroImage. 2014;95 :232–247. doi:10.1016/j.neuroimage.2014.03.034 24657355
75. Smith SM , Miller KL , Salimi-Khorshidi G , Network modelling methods for FMRI. NeuroImage. 2011;54 (2 ):875–891. doi:10.1016/j.neuroimage.2010.08.063 20817103
