University of Southern California Dissertations and Theses

Relationships between lifetime chronic stress exposure, vascular risk, cognition, and brain structure in HIV

(USC Thesis Other)

Relationships between lifetime chronic stress exposure, vascular risk, cognition, and brain structure in HIV

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright 2021 Elissa Charney McIntosh

RELATIONSHIPS BETWEEN LIFETIME CHRONIC STRESS EXPOSURE, VASCULAR
RISK, COGNITION, AND BRAIN STRUCTURE IN HIV

by

Elissa C. McIntosh

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

August 2021

ii
Dedication

This dissertation is dedicated to my partner, John Hemmingsen, who has been by my side
throughout my entire graduate school education. His love and support have been instrumental in
completing this dissertation and the entire doctoral journey. Thank you for believing in me, and
supporting my dreams.

iii
Acknowledgements

I would like to thank my committee, Dr. April Thames, Dr. Richard John, Dr. Beth
Meyerowitz, Dr. Antoine Bechara, and Dr. Mara Mather for their feedback since the inception of
this project. I am especially grateful for Dr. Thames for her mentorship in research, clinical
pursuits, and professional growth during my time with her at USC. I am undoubtedly a better
scientist from working with her. I would also like to thank Dr. Chris Beam for his valuable
guidance and consultation regarding statistics during this project and beyond.
I am also grateful for my friends, colleagues, and lab mates in graduate school both
within and outside of USC that have helped to support me both intellectually and emotionally in
my journey to complete this degree. They have helped me grow as a both a researcher and
clinician. I would especially like to thank Anna Blanken, Aimee Gaubert, Alaina Gold, Lisa
Graves, Mona Khaled, Zoe Mestre, Stephanie Oleson, Lindsay Rotblatt, Kayla Tureson, and
Belinda Yew.
I would also like to thank my mother who cultivated my value for education, and
provided moral and emotional support throughout my life, and especially in the last few years.
Finally, I want to thank the staff in the Social Neuroscience in Health Psychology Lab for
helping me complete this project through data collection and database management.

iv
Table of Contents
Dedication ....................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Tables ...................................................................................................................................v
List of Figures ................................................................................................................................ vi
Abstract ......................................................................................................................................... vii
Chapter 1. Introduction ....................................................................................................................1
Gaps in the Literature .........................................................................................................11
Current Study .....................................................................................................................14
Chapter 2. Methods ........................................................................................................................16
Chapter 3. Results ..........................................................................................................................28
Chapter 4. Discussion ....................................................................................................................37
Future Directions ...............................................................................................................53
Strengths & Limitations ....................................................................................................54
Conclusion .........................................................................................................................56
References ......................................................................................................................................58
Appendices .....................................................................................................................................87
Tables .................................................................................................................................87
Figures................................................................................................................................94
Supplementary Materials .................................................................................................102

v
List of Tables

Table 1. Dimensions of life stress assessed by the Stress and Adversity Inventory for Adults
(Adult STRAIN). ...........................................................................................................................87
Table 2. Demographic and clinical characteristics of baseline study sample. ..............................88
Table 3. Vascular risk clinical data. ..............................................................................................90
Table 4. Demographic and clinical characteristics of participants who completed MRI
evaluation. ......................................................................................................................................91
Table 5. Demographic and clinical characteristics of participants who completed follow-up visit.

........................................................................................................................................................92
Table 6. Relationships between CCSE and brain volumes in the PFC. ........................................93
Supplemental Table 1. Comparison of baseline and MRI-subset samples. ...............................102
Supplemental Table 2. Comparisons of baseline and follow-up samples .................................103
Supplemental Table 3. Relationships between CCSE and individual MRI ROIs .....................104
Supplemental Table 4. Associations of different lifetime stress indices with cognition and PFC
volume. .........................................................................................................................................106

vi
List of Figures

Figure 1. Conceptual Diagram of Primary Aims. .........................................................................94
Figure 2. Relationship between CCSE and Global Cognition. .....................................................95
Figure 3. Relationship between CCSE and Processing Speed. .....................................................96
Figure 4. Relationship between CCSE and Executive Functioning. .............................................97
Figure 5. Relationship between CCSE and PFC Volume. ............................................................98
Figure 6. Relationship between CCSE and OFC Volume. ...........................................................99
Figure 7. Relationship between CCSE and dlPFC Volume. .......................................................100
Figure 8. PFC Volume Mediates Relationship between CCSE and Processing Speed... ...........101

vii
Abstract

Objective: Despite considerable research on stress and brain structure and function, there is
limited knowledge of the effects of chronic stress and lifetime stress exposure on cognition and
brain volumes in humans. Further, despite well-known relationships between stress and vascular
risk, there has been little examination of how vascular risk may mediate or moderate
relationships between chronic stress and cognition. Stress exposure and vascular risk are elevated
in people living with HIV (PLWH), which has important implications for cognitive dysfunction
and decline in this population. The current study seeks to extend the current literature through
examination of relationships between cumulative lifetime exposure to chronic stress (herein
referred to cumulative chronic stress exposure [CCSE]), vascular risk, cognition, and brain
volume in a diverse sample of PLWH and HIV-uninfected (HIV-) adults.
Methods: Participants included 161 PLWH (n = 65) and HIV- (n = 96) community-dwelling
adults. Participants completed the Stress and Adversity Inventory (STRAIN) to measure CCSE,
in addition to measures of recent mood, psychiatric illness, and substance use. To measure
cognition, participants underwent a neuropsychological assessment at baseline and at 2-year
follow-up (n = 79, 49.4% PLWH). Vascular risk was measured using physical examination, and
self-reported medications and medical diagnoses. Vascular age discrepancy, a measure of
cumulative vascular risk, as well as individual vascular risk factors were used to quantify
vascular risk. Regional brain volumes were derived from T1-weighted MP-RAGE images
processed through automated Freesurfer pipeline on a subset who underwent MRI (n = 91;
47.3% PLWH).
Results: Linear regression analyses showed lifetime CCSE was inversely associated with global
cognition (t = -2.381, p = 0.019), which was driven by lower processing speed (t = -3.014, p =
0.003), and executive functioning performance (t = -2.088, p = 0.039). There were no main or
interactive effects of HIV status (p’s > 0.1). CCSE was inversely associated with prefrontal
cortex (PFC) volume (t = -3.751, p = 0.007), including the orbitofrontal cortex (t = -2.727, p =
0.008) and dorsolateral PFC (t = -2.274, p = 0.026), but not subcortical structures associated with
stress/emotion regulation (p’s > 0.1). HIV status did not modify relationships between CCSE and
brain volume; however, PLWH had lower PFC volume relative to the HIV- group (t = -2.099, p
= 0.039). Mediation analysis demonstrated that the relationship between CCSE and processing
speed was mediated by PFC volume (95% confidence interval (CI) = -0.08, -0.00). CCSE was
positively associated with change in learning (t = 2.664, p = 0.010). Vascular risk was unrelated
to CCSE, and did not attenuate relationships between CCSE and cognition. There were no
significant interactions between CCSE and vascular risk on cognition.
Conclusion: This study demonstrates important relationships between CCSE and the brain in
PLWH and HIV- adults. Further, the results suggest that the negative impact of cumulative stress
exposure on cognition, particularly abilities reliant on frontal lobe integrity, may be in part
driven by smaller volumes in the PFC regardless of HIV status. Vascular risk did not attenuate
cognitive findings suggesting that chronic stress exerts direct effects on the brain independent of
vascular risk. Further, vascular risk did not moderate relationships between CCSE and cognition.
These findings emphasize the importance of lifetime stressor exposure, especially chronic stress
exposure, when examining relationships between stress and brain function. This work advocates
for identification of psychosocial interventions to combat chronic stress-related impacts on brain
function in adults at high risk for elevated stressor exposure and premature cognitive aging.
1
Introduction
HIV and Cognition
The advent and widespread use of combination antiretroviral therapy (cART) in the past
two decades has dramatically reduced rates of mortality due to AIDS, and the average life span
expectancy for PLWH treated with cART is nearly comparable to those without HIV. However,
cognitive impairment remains prevalent despite viral suppression with cART. In the United
States, HIV-associated dementia has become rare, nevertheless, milder forms of HIV-associated
neurocognitive disorder (HAND) persist (Heaton et al., 2010). The persistence of neurocognitive
impairment in the context of HIV is important as these deficits affect quality of life and everyday
functioning (Heaton et al., 2004; Thames et al., 2013). Understanding contributions to HAND is
especially important now given the aging HIV population. It is estimated that at least 50% of
patients diagnosed with HIV in the United States are age 50 and older (Kirk & Goetz, 2009).
Studies of cognitive aging in this cohort suggest that HIV may accentuate cognitive aging (Cole
et al., 2017; Kuhn et al., 2018), thus it is important to understand the mechanisms that underlie
the persistence of cognitive dysfunction and decline in this population.
HIV and Stress
PLWH are disproportionately members of minoritized groups (e.g., racial and sexual
minority), of low-socioeconomic status and have higher rates of stressful life events, and trauma
compared to the general population (Cargill & Stone, 2005; Machtinger et al., 2012). Living with
HIV is associated with several stressors that are directly (e.g., strict medication schedule,
medication side effects) and indirectly (e.g., stigma, interpersonal difficulty, emotional issues)
related to HIV (Mitchell & Linsk, 2004; Pakenham, 2010). Increased stress in PLWH has
important implications for brain and immune functioning. Stress and psychological factors (e.g.,
2
social inhibition) may negatively affect disease progression, treatment adherence, and promote
risk taking behaviors (e.g., substance abuse, risky sex) (Cohen et al., 2007; Kaijage & Wexler,
2010).
HIV and Vascular Risk
Cardiovascular disease (CVD) risk factors have emerged as an important potential risk
factor for neurocognitive dysfunction in PLWH in the cART era. It has been demonstrated that
PLWH have higher rates of cardiovascular and cerebrovascular disease than uninfected
individuals (Demir et al., 2018; Singer et al., 2013; So-Armah & Freiberg, 2014). Many HIV
cohort studies have demonstrated that PLWH have a higher risk of acute myocardial infarction or
coronary heart disease (Triant, 2013). Similarly, HIV infection is associated with subclinical
CVD and individual vascular risk factors such as dyslipidemia, type 2 diabetes mellitus (T2DM),
hypertension, and abdominal obesity (Grinspoon & Carr, 2005; Hanna et al., 2015; Seaberg et
al., 2010; Triant, 2013).
The etiology of CVD in HIV is multifactorial. Possible mechanisms for the elevated rates
of CVD in HIV include HIV-specific factors, cART cardiometabolic side effects, traditional
CVD risk factors, and nontraditional CVD risk factors and comorbidities (e.g., Hepatitis C
coinfection; substance use) (McIntosh et al., 2021).
Stress and Central Nervous System Structure and Function
Perceived stress and exposure to stressful life events are widely recognized risk factors
for cognitive dysfunction. Stress is increasingly being recognized as a contributor to cognitive
dysfunction in PLWH (Spies et al., 2020; Valdez et al., 2016). The relationships between stress
and cognitive function have been widely studied, especially in animal studies. In adults, much of
the literature has focused on early life stress exposure, recent stress exposure, and perceived
3
stress in relation to cognition and brain structure. Though operational definitions of stress vary
widely across studies, there is substantive scientific evidence supporting relationships between
experienced stress on cognition and brain structure, particularly among regions that are involved
in stress signaling and emotion regulation.
Stress is posited to negatively affect the brain via glucocorticoid (GC) stress hormones
released by activation of hypothalamic-pituitary-adrenal (HPA) axis (Lupien et al., 2009).
Briefly, HPA axis activation begins with release of corticotropin-releasing factor from the
hypothalamus, which triggers the release of adrenocorticotropin hormone (ACTH) from the
pituitary gland. ACTH then travels into the blood stream where it binds to receptors on the
adrenal glands, which then stimulates release of GCs that cross the blood-brain-barrier and bind
to GC receptors in the brain (Marin et al., 2011). GC receptors are distributed in subcortical and
cortical structures, including the hippocampus, amygdala, and prefrontal cortex (PFC). These
structures are affected by GCs and have effects on their regulation (Lupien et al., 2018). While
the amygdala activates the HPA axis, the hippocampus and PFC have inhibitory functions
(Marin et al., 2011). The GC cascade hypothesis (Sapolsky et al., 1986), now known as the
Neurotoxicity Hypothesis, proposes that chronic stress via high levels of GCs in the brain leads
to dysregulation of HPA-axis resulting in structural brain changes described below and
corresponding declines in cognitive functions such as learning, memory, and executive functions.
Chronic administration of GCs in animal studies, mimicking the effects of chronic stress, result
in changes to dendritic morphology (Liu & Aghajanian, 2008; Wellman, 2001), which may be
mediated by excitotoxicity or oxidative stress (Zhu et al., 2007). Another potential mechanism is
reductions of brain-derived neurotrophic factor, which is important for neuronal survival,
synaptic plasticity, and memory (Gourley et al., 2009).
4
In animal studies, high exposure to GCs is associated with hippocampal neuronal loss,
dendritic atrophy, and decreased hippocampal volume, and decreased neurogenesis in the dentate
gyrus (Marin et al., 2011). Longitudinal studies have shown relationships between cortisol levels
and chronic stress and atrophy in the hippocampus and orbitofrontal cortex (Gianaros et al.,
2007; Lupien et al., 1994; Lupien et al., 1998). Chronic stress also causes structural alterations to
the PFC. Animal studies show chronic stress induces architectural changes and reductions of
dendrite length, branching and spine density (Holmes & Wellman, 2009; Radley et al., 2006).
Changes to the PFC in response to stress are thought to occur more quickly than the
hippocampus (Brown et al., 2005). Interestingly, animal studies show stress-related changes in
the PFC may be reversible in youth, but less so in older animals (Bloss et al., 2010). Chronic
stress has also been shown to affect the relationship between the PFC and hippocampus, which is
important in memory consolidation (Cerqueira et al., 2007). The amygdala, in contrast to
changes in the hippocampus and PFC, expands as a result of chronic stress (Vyas et al., 2002).
The pattern of changes in these structures results in loss of dendritic material in structures
providing negative feedback in stress response (e.g. hippocampus, PFC), and growth of dendritic
material in the amygdala which functions to heighten stress response (Arnsten, 2009).
Early Life Stress
There are multiple conceptualizations of stress in the literature that differ based on time
of exposure (e.g., early life versus recent) and type of stress exposure (e.g., acute versus chronic),
which are theorized to differential affect CNS function and structure.
Early life stress has detrimental impacts on cognition and brain structure that may persist
into adulthood (Hedges & Woon, 2011). Animal studies have shown maternal deprivation is
associated with enduring dendritic reduction in the PFC (Pascual & Zamora-León, 2007). The
5
amygdala appears to be more associated with stress in early life, while hippocampus
abnormalities are typically seen later in childhood. The amygdala is hypothesized to precede
hippocampal development and therefore may be more vulnerable to early life stress whereas the
hippocampus may be more vulnerable in later years (Tottenham & Sheridan, 2009). In human
studies, early life stress has been associated with enlarged amygdalae, decreased corpus callosum
integrity, and smaller hippocampi, anterior cingulate cortex, and caudate in children/adolescents
and adult samples (Andersen et al., 2008; Clark et al., 2012; Cohen et al., 2006; Mehta et al.,
2009; Paul et al., 2008; Seckfort et al., 2008; Stein et al., 1997; Tottenham et al., 2010).
Interestingly, one study showed age of childhood abuse was differentially associated with
reductions in different brain volumes in early adulthood suggesting brain regions are uniquely
impacted depending on age of exposure (Andersen et al., 2008). Similarly, early life stress was
associated with decreased white matter integrity in middle-aged to older adults but not younger
adults (Seckfort et al., 2008).
Importantly, studies show stress, especially early life stress, to be particularly harmful to
cognitive functioning and brain structure in PLWH. In a study of PLWH and demographically
similar controls, participants who reported at least three early life adverse life-events on the
Early Life Stress Questionnaire performed worse on processing and psychomotor speed. Post-
hoc analyses showed that for both HIV status groups (infected vs. uninfected), those who
reported high levels of early life stress groups performed worse in comparison to the HIV-
uninfected individuals who reported low levels of early life stress. Interestingly, there were no
significant differences in processing and psychomotor speed performance as a function of early
life stress in PLWH. However, when examining brain volume, there was an interaction between
HIV-status and early life stress on amygdala volume. Specifically, PLWH with high early life
6
stress demonstrated larger amygdala volume as compared to HIV-uninfected early life stress
groups and PLWH with low early life stress. In PLWH, increased volume of the amygdala was
associated with higher scores on a measure of early life stress, lower nadir CD4, and worse
psychomotor/processing speed (Clark et al., 2012). In the same cohort, PLWH with high early
life stress showed increased reaction time intra-individual variability, a behavioral marker
associated with brain volume and neuropsychiatric symptoms, and lower gray and white matter
volume relative to PLWH with low early life stress (Clark et al., 2018). Importantly, findings in
the Clark et al., papers remained significant after accounting for current depressive symptoms
and perceived stress. In another cohort, women living with HIV with childhood trauma, as
defined by history of emotional neglect, physical neglect, physical abuse or sexual abuse, tended
to perform worse than HIV-uninfected women without childhood trauma and control groups
during baseline on most neurocognitive domains including learning, memory,
processing/psychomotor speed, executive functioning (e.g., inhibition, perseveration), and
language. Longitudinal analyses showed an interaction between HIV status and childhood
trauma history such that those with HIV and childhood trauma history unexpectedly showed
improvement over time on novel problem solving and conceptualization, and semantic verbal
fluency, which may be attributed to increased cART use in this group over follow-up (Spies et
al., 2017). However, this study was limited in that they did not measure or account for recent
depressive symptoms or stress exposure history after childhood. Together these findings indicate
that the adverse effects of early life stress exposure on cognitive dysfunction may be more
prominent among PLWH, while more research is needed on cognitive change.
Recent Stress
7
Studies also frequently focus on recent stress as defined by perceived life stress (e.g.,
Perceived Stress Scale) or self-report life event checklist measures. In addition to studies of early
life stress among PLWH, studies of perceived stress in the last month and recent exposure to
stressful events as it relates to cognition have also been conducted (Pukay-Martin et al., 2003;
Rubin et al., 2015). One study showed exposure to stressful events in the past year was
associated with worse cognitive function in PLWH, but not HIV-uninfected participants in an
all-male sample (Pukay-Martin et al., 2003). However, this study had important limitations such
that they did not measure any previous stress exposure or analyze interactions between HIV
status and stress exposure. Further, the PLWH group was substantially larger than the HIV-
uninfected group (n = 251 vs n = 82) and did not report effect sizes which makes comparisons of
these relationships difficult. In another study, perceived stress in the past month was inversely
associated with attention, verbal learning, processing speed, and executive functioning regardless
of HIV status, whereas there was an interaction between HIV-status and verbal memory such
that the relationship between perceived stress and verbal memory was only significant among
women with HIV (Rubin et al., 2015). Notably, the study did not measure or control for any past
history of stress exposure. Neuroimaging studies in adults have shown recent perceived stress is
associated with lower frontal brain volumes, parahippocampal, and hippocampal volume in both
PLWH and non-infected populations (L. H. Rubin et al., 2016; Zimmerman et al., 2016).
Studies of perceived stress in the last month using the perceived stress scale have
demonstrated stress is associated with cognitive decline in older adults (Aggarwal, Wilson, et al.,
2014; Munoz et al., 2015; Turner et al., 2017). However, perceived stress was not associated
with baseline cognition or decline in a sample of adults with amnestic mild cognitive impairment
(Sussams et al., 2020). Notably, the majority of studies on perceived stress and cognitive decline
8
has been performed in older adults, which may be important as it is posited that the negative
effects of stress on the body and cognition may be more salient in older adults (Lupien et al.,
2009).
Chronic Stress
There is less research on relationships between chronic stressors and cognition among
adults in humans, and the few studies tend to focus on isolated stressors (e.g., caregiving). Most
studies on chronic stressors have relied on behavioral stress manipulations in rats and rodents.
Chronic elevations in cortisol are shown to predict cognitive dysfunction in attention and
memory processing and reduced hippocampal volume in healthy older adults (Lupien et al.,
1994; Lupien et al., 1998). In a study of chronic caregiver stress, higher perceived burden in
caregivers was associated with worse attention and processing speed in caregivers, but not
controls (Caswell et al., 2003). In outpatient adults treated for chronic stress, patients performed
worse than controls on verbal learning, set-shifting, and phonemic fluency (Ohman et al., 2007).
Findings of chronic stress-related deficits in attention and memory are consistent with research
findings among people with post-traumatic stress disorder (Brandes et al., 2002; Sumner et al.,
2017).
Stressor Exposure and Cognitive Decline
In general, the literature of cumulative stressor exposure and cognitive decline is limited
and shows mixed findings depending on cognitive status and participant demographics. One
study showed no relationship between cumulative stress exposure over the last three years and
cognitive decline in older adults, however some chronic stressors (e.g., illness of partner or
relative or serious conflicts) were associated with better cognitive function at follow-up (Comijs
et al., 2011). In another study, greater event-based stress ratings were associated with faster
9
decline in participants with mild cognitive impairment but not in cognitively normal participants
(Peavy et al., 2009). Another study showed the relationship between stressful life events over the
last three years and cognitive decline in a non-demented elderly population was modified by both
years of education and age such that those with less education and younger participants were
more susceptible to stressor-related cognitive decline (Tschanz et al., 2013). The majority of
other studies focused on stressor exposure have looked at particular stressful life events and
cognitive declines. For example, several studies have associated with death of a spouse is
associated with cognitive decline or cognitive impairment in older adults (Aartsen et al., 2005;
Xavier et al., 2002). Similarly, another study showed midlife job strain was associated with
future cognitive decline (Andel et al., 2011). The fact that some studies have showed that
individual stressor exposures, but not cumulative exposures, are related with cognitive
performance and change suggest that individual stressors may have differential impact on
cognition and trajectories (Comijs et al., 2011; Rosnick et al., 2007). Alternatively, it may
suggest that studies assessing cumulative stressor exposure have used measures that do not query
enough stressors to demonstrate a cumulative effect.
Stress and Vascular Risk and Cardiovascular Disease
In addition to the direct adverse effects of chronic stress on the brain, exposure to
stressors and psychological stress may also indirectly affect the brain and its functioning through
the development of vascular risk factors including CVD and cerebrovascular disease. In
adulthood, chronic stress burden, as measured by cumulative current stressors of at least 6-month
duration was associated with higher prevalence of coronary heart disease, stroke, diabetes, and
hypertension (Gallo et al., 2014). Similarly, stressful life event exposure over the past year was
associated with the metabolic syndrome and its individual components (e.g., insulin resistance,
10
obesity, elevated triglycerides) in a large population-based study (Pyykkönen et al., 2010).
Longitudinal studies have also reported stressful life events predict future vascular risk,
including the metabolic syndrome; one study reported extremely stressful life events in middle
aged predicted development of the metabolic syndrome 15 years later (Räikkönen et al., 2007).
Perceived stress, or psychological distress, has also been associated with vascular risk factors
and outcomes such as high blood pressure (Wiernik et al., 2014), incident stroke and death
(Everson et al., 2001; Henderson et al., 2013; Truelsen et al., 2003), atrial fibrillation (O'Neal et
al., 2015), cerebral infarction and lower brain volume (Aggarwal, Clark, et al., 2014), and
smoking (Gallo et al., 2014).
Longitudinal studies suggest relationships between stress and cardiovascular disease are
mediated via lifestyle behaviors (e.g., physical inactivity, smoking, sedentary lifestyle, alcohol
use) and development of biological risk factors (e.g., hypertension, diabetes) (Rod et al., 2009).
Notably, stress may be more related to coronary heart disease and mortality in low
socioeconomic status (SES) groups (Redmond et al., 2013). Elevated stress may increase risk of
vascular risk factors and stroke through dysregulation of glucocorticoids via the HPA axis,
changes in cardiac autonomic tone via sympathetic-parasympathetic mechanisms, and
inflammatory processes (Rozanski et al., 1999). Elevated cortisol associated with chronic
stressor exposure can increase blood pressure and levels of cholesterol, triglycerides, and glucose
in blood which in turn increase risk for CVD and cerebrovascular disease (Whitworth et al.,
2005). Stress-related activation of the sympathetic nervous system and HPA axis may also more
directly increased risk for CVD through atherosclerosis; these stress responses can cause
endothelial dysfunction which is a driver of atherosclerosis development. Further, stress may
11
accelerate atherosclerosis progression through chronic low-grade inflammation (Lagraauw et al.,
2015).
Vascular Risk and Cognition
The relationship between stress and vascular health may have important implications for
cognitive functioning, especially in PLWH given elevated exposure to stressors and increased
vascular risk. Vascular risk factors, including hypertension and diabetes, are related to cognitive
dysfunction and cognitive decline in PLWH and uninfected populations (Crystal et al., 2011;
Duron & Hanon, 2008; McIntosh et al., 2021), and risk for dementia and Alzheimer’s disease
(Cechetto et al., 2008; Jefferson et al., 2015). Both biological risk factors (e.g., obesity,
hypertension, T2DM) and behavioral risk factors (e.g., physical inactivity, smoking) can
adversely affect cognition and cognitive aging. Individual risk factors and cumulative risk factors
are both linked to cognitive decline in late-middle aged and early older aged adults (Dregan et
al., 2013; Knopman et al., 2001). Vascular risk factors predominantly affect frontal-subcortical
white matter tracts, producing deficits in processing speed, learning/memory, executive
functions, and psychomotor speed (Cohen et al., 2009).
Gaps in the Literature
Despite the wealth of research on stress and the brain, the majority of studies in human
subjects vary greatly with their definition of stress, and are largely limited to measures of time-
limited psychological responses to stressors (e.g., perceived stress in the last month, feeling
distressed) or time-limited stressor exposure (e.g., early life stress, recent stressors in the past
month to year). Further, a majority of stress measures consists of check lists or Likert ratings that
do not fully capture the impact of the event (e.g., subjective appraisal, duration). In the stress
literature, there is consensus that exposure to stress over the course of one’s lifetime produce a
12
cumulative or additive effect on biological processes that increase risk for disease (Geronimus,
1992; Lupien et al., 2009). As such, many contemporary models of stress and health posit that
stress exposure across the lifetime affects health outcomes (Graham et al., 2006; Lupien et al.,
2009). When the stress response is activated often and prolonged, biological “wear and tear” or
allostatic load occurs that increases risk for disease. Thus, according to this hypothesis, the
examination of cumulative stressor exposure is central to health and supports the importance of
examination of the cumulative effects of stressor exposure over the lifetime on cognition and
brain structure.
In addition to the paucity of research assessing stressor exposure over long periods, the
current literature is limited in that that there is little differentiation between type of stress
exposure (e.g., acute vs. chronic). There is evidence that exposure to chronic difficulties is
particularly detrimental to health due in part to systemic low-grade inflammation (Cohen et al.,
2007; Slavich & Irwin, 2014). According to Cohen, “As exposure persists, there are increased
probabilities of the stressor being present at points of vulnerability in the disease process; of
long-term care or permanent changes in the emotional, physiological, and behavioral responses
that have downstream influences on disease (Cohen et al., 2007); and of increased wear and tear
on the body (e.g., allostatic load) (McEwen, 2004)” (Cohen et al., 2019). Further, most measures
that query about stressful life events do not include a subjective appraisal of the experience. This
is an important limitation as the perception of stress is posited to drive the relationship between
stressful events and adverse health outcomes (Lazarus & Folkman, 1984). Consistent with this,
one study showed negative appraisal of events, rather than frequency, was related to cognitive
performance (Rosnick et al., 2007). Subjective appraisal of stress experience (i.e., stressor
severity) is particularly important to measure as it takes into account both exposure and
13
perception of the experience. Lastly, information about duration of stressor exposure is typically
omitted; however, duration is important in establishing the chronicity of a stressor and its
potential impact on health and behavior.
Taken together, relationships between perceived severity of lifetime cumulative chronic
stress exposure (herein referred to as CCSE) and cognition and the role of brain integrity
represent an important gap in the literature in HIV-uninfected populations as well as in PLWH.
We currently have very little knowledge of how lifetime stress exposure influences the brain in
adult populations. While contemporary models of stress suggest that lifetime cumulative stress
should negatively impact cognition and brain structure, this has not been tested empirically likely
due to previous scarcity of stressor measures able to capture lifetime stress in a thoughtful and
efficient manner. Addressing this gap in the literature is critical in furthering our understanding
of the impact of lifetime stress on health, and brain function more specifically. Additionally,
there have been very few studies documenting relationships between chronic stress and cognitive
decline, which may have important implications for aging in PLWH.
Another critical gap in the literature is the lack of a focused look at relationships between
vascular risk, stress, and cognition. While there have been a few studies that have incorporated
vascular risk factors as covariates when examining relationships between stress and cognition to
prove independent effects of stress, no studies to our knowledge have looked relationships
between stress indices and vascular risk in these studies. For example, two studies on recent
perceived stress have demonstrated relationships with cognition independent of vascular risk
factors, but did not look at relationships between perceived stress and vascular risk, or vascular
risk as moderators in cognitive analyses (Aggarwal, Wilson, et al., 2014; Turner et al., 2017).
One study investigated whether vascular risk factors moderated relationship between recent
14
stressful life events and white matter hyperintensities, but not cognition (Johnson et al., 2017).
Thus, despite established relationships between stress and cognition, vascular risk and cognition,
and stress and vascular risk, few studies have examined mediating and moderating relationships
between these constructs. Examination of these relationships may be particularly important in
PLWH as they have elevated rates of stress exposure and vascular risk factors. Better
understanding of relationships between stress and vascular risk on brain structure and function is
critical given the ubiquity of these cognitive risk factors.
Current Study
In part one of the dissertation, we aimed to build off existing literature by examining
relationships between CCSE, cognition, and brain volume in a sample of community-dwelling
PLWH and HIV-uninfected (HIV-) adults. In brief, while there is a robust literature on stress and
cognition, there are several important limitations and gaps in the extant literature. First, there are
very few studies that take a systematic approach to measuring cumulative life stress in adult
populations, and the majority of studies are limited to specific time frames (e.g., recent stress
within the past month to year, early life stress). Second, studies vary in their definition of stress
(e.g., psychological distress vs. stressful event exposure), which represent distinct phenomena.
Third, of those studies that do capture stressful event exposure, some do not include a subjective
appraisal of the impact of the stressor exposure. Fourth, studies often do not distinguish between
acute stress and chronic stress, which may have differential impacts on health. To address the
aforementioned limitations of prior stress research, our study utilized the Stress and Adversity
Inventory (STRAIN; Table 1), a lifetime measure of stressor exposures that captures severity,
frequency, timing, duration, and type of stressor (e.g., acute versus chronic). The STRAIN has
15
been associated with various psychological and biological processes and disorders (Shields &
Slavich, 2017; Slavich & Shields, 2018).
In part two of the dissertation, we incorporated measures of vascular risk in primary aims
and investigated relationships between vascular risk and CCSE and cognition, and mediating and
moderating effects of vascular risk on relationships between CCSE and cognition. To accomplish
this, we defined vascular risk using vascular age discrepancy, a cumulative vascular risk burden
score derived from the Framingham Heart Study 10-year CVD risk algorithm, as well as
presence of individual vascular risk factors (e.g., hypertension, T2DM, dyslipidemia, obesity,
current smoking).
We set to explore the following aims and hypotheses in the dissertation:
Aim 1. Examine relationships between CCSE, cognitive function, and brain structure.
Hypothesis 1a: CCSE will be negatively associated with global cognition and processing
speed, executive functioning, learning, and memory.
Hypothesis 1b: CCSE will be negatively associated with volumes in the PFC,
hippocampus, and insula, and positively associated with amygdala volume.
Aim 2. Examine the mediating role of brain volume in the relationship between CCSE and
cognition.
Hypothesis 2: Brain volume will mediate significant relationships between CCSE and
cognition.
Aim 3: Examine if CCSE is related to cognitive change.
Hypothesis 3: Increased CCSE will be associated with cognitive decline from baseline to
2-year follow-up.
16
Aim 4. Examine relationships between CCSE, vascular age discrepancy, individual vascular risk
factors, and cognition.
Hypothesis 4a: Increased CCSE will be associated with increased vascular age
discrepancy, and increased rates of individual vascular risk factors.
Hypothesis 4b: Vascular age discrepancy and presence of individual vascular risk factors
will be negatively associated with cognition such that increased vascular age discrepancy
and presence of individual vascular risk factors are associated with worse cognitive
performance.
Aim 5. Examine if vascular age discrepancy mediates or moderates the relationship between
CCSE and cognition.
Hypothesis 5a: If vascular risk discrepancy is associated with both CCSE and cognition,
vascular risk will mediate the relationship between CCSE and cognition.
Hypothesis 5b: Vascular age discrepancy and CCSE will have an interactive effect on
cognitive outcomes such that the negative relationship between CCSE and cognition will
be stronger at high levels of vascular risk.
Methods
Participants
Participants included in this study were part of a larger study called the Lifetime Stress
and Aging Study (RO1 MH114761; PI: April Thames). For the parent study, participants were
recruited from HIV clinics and the local community through local advertisements as well as
participant word of mouth. HIV status was confirmed via serologic testing (i.e., Western Blot
confirmed by ELISA). Questionnaires and screeners about medical, neurological and psychiatric
history were used to screen for potential confounds. Briefly, participants were screened for
17
neurological, psychiatric, illicit drug use, current substance abuse (excluding alcohol and
marijuana), and medical confounds using the Structured Clinical Interview (SCID) for DSM-IV
(First et al., 1995), Mini-Mental Status Exam (MMSE) (Folstein et al., 1975), urine toxicology
test, and questionnaires about neurological and medical history. Participants were excluded
based on history of head injury with loss of consciousness (>30 minutes), neurological disease
(e.g., seizure disorder), and likely cognitive impairment (e.g., MMSE < 26).
The present study used a subset of participants who had available data for all primary
variables of interest. Participant data was also screened for additional exclusionary criteria which
included the following: age < 30 or age > 74 years-old, history of CVD (e.g., coronary artery
disease, atrial fibrillation), cardiac event (e.g., myocardial infarction), or cerebrovascular disease
(e.g., stroke) (N=6), history of HIV-related opportunistic infections (e.g., meningitis) (N=2), and
body mass index <18.5 due to limited number of underweight individuals (N=3), and its known
relationship to cognitive dysfunction (Sabia et al., 2009).
After applying criteria for data availability and additional exclusions (as described
above), 161 participants (65 PLWH and 96 HIV-uninfected [HIV-] controls) were eligible for
analysis. Participants all had completed baseline neuropsychological assessment, medical history
and assessment (which included assessment of vascular risk), and psychosocial assessments
(e.g., interview and self-report questionnaires on stress and mood). A subset of individuals
returned for a follow-up visit (n = 79) and completed neuropsychological testing, and
psychosocial questionnaires. The average time to follow-up was 2.21 years (SD = 1.01).
Demographics
Demographic data of age, education, racial/ethnic identity, and sex were obtained through
self-report (Table 2).
18
Socioeconomic Status
Socioeconomic status (SES) was measured using the Hollingshead Four-Factor Index of
Social Status; the Hollingshead scale algorithm utilizes information about occupational prestige,
income, sex, and marital status (Hollingshead, 1975). Hollingshead scores ranged from 14 to 66.
In addition, for descriptive purposes only, data was previously collected about total household
income, and average household income per family member.
Lifetime Cumulative Chronic Stressor Exposure
The Stress and Adversity Inventory (STRAIN) was used to measure lifetime cumulative
chronic stressor exposure severity (CCSE) (Slavich & Shields, 2018). The STRAIN asks about
exposure to various stressors over the lifetime including gestation period and birth. For each
stressor endorsed, participants then complete follow-up questions regarding severity, frequency,
timing, and duration of the specific stressor. Stress is measured in 11 domains and 5
characteristics (see Table 1). Across stressor domains, the STRAIN computes data for total count
and severity of all types of stressors, and acute life events (N = 26), and chronic difficulties (N =
29) separately. Severity is calculated by adding severity ratings for each stressor endorsed. Data
is separated into early adversity (e.g., childhood, <18 years-old) versus adulthood stressors.
Number of stressors and severity of stressors are available for each type of domain. The STRAIN
has demonstrated good predictive validity for physical and cognitive health outcomes. For
example, lifetime stressor total count was associated with self-reported physical and mental
health complaints, sleep difficulties over the past month, worse inhibitory control, and greater
doctor-diagnosed general health problems and autoimmune disorders (Cazassa et al., 2020)
For the purposes of the present study, the variable of interest was lifetime cumulative
chronic stressor exposure severity (referred to as CCSE) (range: 0 to 106). Additional variables
19
from the STRAIN examined included childhood CCSE (range: 0 to 41), and adulthood CCSE
(calculated by subtracting childhood CCSE from lifetime CCSE; range: 0 to 84). Further, total
stress severity related to health/treatment (range: 0 to 35) was utilized as a covariate adjust for
HIV-related differences in stress exposure related to health/treatment.
Psychiatric Measures
Depressive symptoms were measured using the Beck Depression Inventory – version 2
(BDI-2) (Beck et al., 1996). The BDI-2 is self-report questionnaire that measures severity of
depressive symptoms. Respondents are instructed to rate 21 groups of statements related to
depressive symptoms (e.g., feelings of guilt, failure, loss of pleasure, etc.) based on how they
have been feeling in the past two weeks. It is used widely for both research and clinical purposes.
Vascular Risk Assessment
Vascular risk factors were determined with physical examinations, clinical interviews,
and medication review. Seated brachial artery blood pressure, weight, and height were measured
as part of the physical exam. Body mass index (BMI) was calculated as weight (kg) divided by
height (meters) squared. Diagnosis and treatment history for hypertension, T2DM, dyslipidemia,
and CVD were documented via interviews and medication review. Current smoking was defined
as smoking tobacco in the last 30 days and a lifetime history of at least 100 cigarettes. The
Framingham Heart Study (FHS) CVD 10-year risk algorithm was used to quantify vascular risk
burden (D'Agostino et al., 2008). The algorithm uses the following variables to predict both risk
of developing CVD in the next 10 years, and estimate vascular age: age, systolic blood pressure,
hypertension pharmacotherapy, presence of T2DM, BMI, current smoking, and gender. The FHS
CVD 10-year risk score has been associated with cognition and cognitive decline in both PLWH
and HIV- populations (Chow et al., 2020; Wright et al., 2015). Similarly, vascular age has been
20
associated with worse cognitive performance in middle-aged and older adults and progression
from mild cognitive impairment to dementia in older adults (Badran et al., 2019; Viticchi et al.,
2017). Body mass index was used in lieu of fasting total cholesterol and HDL cholesterol given
this data was not collected in majority of participants (N = 34 with data), and is suitable proxy
for fasting cholesterol based on previous research (Faeh et al., 2012). In the present study, we
used derived vascular age to calculate vascular age discrepancy (i.e., difference between actual
age and predicted vascular age based on algorithm).
Formula: Vascular Age Discrepancy = Chronological Age – Predicted Vascular Age
A negative value indicates predicted vascular age is greater than actual age, while a
positive value indicates predicted vascular age is less than actual age. Higher negative values
indicate worse health. Vascular age discrepancy was chosen in lieu of 10-year CVD risk score
(e.g., percent risk of developing CVD in 10 years) due to fact that it is age normed; this is
important in the sample because our sample encompasses a wide age range.
In addition to calculating 10-year CVD risk and vascular age discrepancy, participants
were evaluated for the presence of hypertension, dyslipidemia, and obesity. Hypertension was
defined using established clinical guidelines for stage 2 hypertension (systolic blood pressure
≥140 and/or diastolic blood pressure ≥90 mm Hg), history of antihypertensives, or self-reported
hypertension diagnosis. Dyslipidemia was defined using medication history (e.g., statin use), or
self-reported elevated total and/or LDL cholesterol, and/or low HDL cholesterol. Obesity was
defined as BMI ≥30.0 kg/m
2
. Definitions for current smoking and T2DM are described above.
Neuropsychological Assessment
Participants underwent a comprehensive neurocognitive assessment battery at baseline
that assessed the following domains: attention and information processing speed (Wechsler
21
Adult Intelligence Scale-4
th
Edition [WAIS-IV]-Coding and Symbol Search; Trail Making Test-
Part A; and Stroop Interference Test-Color and Word Naming), verbal fluency (Controlled Oral
Word Association Test-FAS and Animal Fluency), learning and memory (Hopkins Verbal
Learning Test-Revised [HVLT-R], Total Learning, Delayed Recall; Brief Visuospatial Memory
Test-Revised [BVMT-R], Total Learning, Delayed Recall), executive functioning (Trail Making
Test-Part B; Stroop Interference Test-Interference; and WAIS-IV-Letter-Number Sequencing),
and motor function (Grooved Pegboard) (Benedict et al., 1998; Benedict et al., 1996; Golden &
Freshwater, 2002; Ruff et al., 1996; Ruff & Parker, 1993; Wechsler, 2008). Raw test scores were
converted into age-adjusted T scores for baseline cross-sectional analyses. Global and individual
cognitive domain scores were computed by averaging T scores for individual tests (Heaton et al.,
1991; Miller & Rohling, 2001). For longitudinal analyses, z-scores for baseline and follow-up
cognitive data were computed based on whole sample means and standard deviations for each
individual test. Z-scores for individual tests were averaged to create composite z-scores.
For all analyses, global cognition was examined and a priori individual domains:
processing speed (Trails A, Stroop Color, Stroop Word, Symbol Search, Coding), executive
functioning (Trails B, Letter-Number Sequencing, Stroop Interference, FAS), learning (HVLT-R
total recall, BVMT-R total recall), and memory (HVLT-R delayed recall, BVMT-R delayed
recall).
Premorbid estimated intelligence was assessed using a test of irregular word reading
(WRAT).
Structural Neuroimaging
A subset of participants completed neuroimaging assessment (n = 91; demographic and
clinical information is described in Table 3). Imaging data was not yet available from
22
participants from the USC cohort (n = 29). Other participants did not complete neuroimaging due
to MRI contraindications such as presence of metal in the body and/or claustrophobia (n = 36) or
they had too much motion artifact to analyze data (n = 5). Structural magnetic resonance imaging
data was collected on 3T Siemens scanners. The parameters of the T1‐weighted MPRAGE scan
are TR/TE/TI = 2500/1.81/1000ms, 8° flip angle, 1.0 × 1.0 × 1.0 mm
3
resolution, acquisition
time 5:13 min. These structural scans were processed using the FreeSurfer image analysis suite
(http://surfer.nmr.mgh.harvard.edu). Standard FreeSurfer automated cortical parcellation processing
procedures were employed (Fischl & Dale, 2000; Fischl et al., 2002). Each subject’s post-
processing outputs were manually inspected for reconstruction accuracy in the Talairach
transform, skull strip, white and pial surfaces, and segmentations. Cortical thickness regions of
interest and volumetric estimates were extracted from automatic surface parcellation labels using
the Desikan/Killiany Atlas (Desikan et al., 2006).
Volumetric region-of-interest (ROI) analyses were chosen based on the extant literature
and included regions associated with emotion and stress regulation networks. ROIs included the
following regions: PFC regions (lateral orbitofrontal cortex [OFC], medial OFC, caudal anterior
cingulate, rostral anterior cingulate, superior frontal, caudal middle frontal, rostral middle frontal,
pars opercularis, pars triangularis, pars orbitalis), limbic system structures (amygdala,
hippocampus), and associated limbic structures (insula). A composite PFC variable was
computed using the average of all individual regions in the PFC specified above, as well as
subregion composites (e.g., ventrolateral PFC (vlPFC), dorsolateral PFC (dlPFC), anterior
cingulate, and OFC). The OFC composite was calculated as the average of lateral OFC and
medial OFC volumes. The anterior cingulate composite was calculated as the average of rostral
anterior cingulate and caudal anterior cingulate volumes. The dlPFC composite was calculated as
23
the average of superior frontal, caudal middle frontal, and rostral middle frontal gyri volumes.
The vlPFC was calculated as the average of pars opercularis, pars triangularis, and pars orbitalis
volumes.
Follow-Up
A subset of individuals returned for an approximate 2-year follow-up (n = 97;
demographic and clinical information is described in Table 4). To attenuate practice effects on
cognitive tests, participants were excluded if their follow-up time was less than 11 months
leaving a total of 79 participants.
Statistical Analysis
SPSS version 26 was utilized for all analyses. The PROCESS macro was used to test all
mediation, moderation, and moderated-mediation analyses (Hayes, 2013).
Prior to analyses, all variables were assessed for outliers and normality (e.g., skewness,
kurtosis). For outlier analysis, dependent variables (e.g., mediator and outcome variables) were
screened by converting raw data to z-scores and inspected them for errors. For brain volumes,
exclusion criteria of z > 3.0 and z < -3.0 were applied to each individual region analyzed (Waters
et al., 2019); 3 outliers for identified for PFC region and 1 outlier was identified for amygdala.
Outliers were removed from all relevant analyses.
Global cognition, processing speed, executive functioning, and all MRI volumes were
normally distributed according to Shapiro-Wilks test (p’s > 0.1). Learning and memory were not
normally distributed (p = 0.002 and p = 0.001), respectively. Log-transformed did not resolve
non-normality, thus unadjusted values were used. For other non-normally distributed variables
(e.g., CCSE, depressive symptoms), parametric testing was used as appropriate (e.g., Spearman
correlations, Mann-Whitney tests).
24
Demographic and Clinical Analyses
ANOVA, Mann-Whitney, and chi-square analyses were used to compare PLWH and
HIV- groups on demographic variables, SES, and psychosocial variables (Table 2). Additional
comparisons were run on those who completed MRI (Table 3) and those who completed follow-
up (Table 4). HIV status group differences were also examined for all vascular risk factors
(Table 5). Comparisons of those who completed MRI versus not and those who completed
follow-up versus not were also examined (Supplemental Table 1 and Supplemental Table 2,
respectively).
Correlates of CCSE
Spearman correlation and Mann-Whitney analyses were run to look at relationships
between CCSE and demographic variables, psychosocial, and psychiatric variables.
Primary Aims Analyses
For aims 1 and 2, the following variables were selected as covariates: age, education,
HIV-status, race, premorbid estimated intelligence (measured by raw WRAT score), recent
depressive symptoms (measured by BDI-II), health-related stress (measured by STRAIN), and
HIV-status by CCSE interaction. Health/treatment stress was included to adjust for HIV status
differences in this domain. Further, intracranial volume was added in MRI analyses. In aim 3,
time to follow-up was also included as a covariate to account for variation in time between
baseline visit and follow-up visit.
Aim 1: Examine relationships between CCSE, cognitive function, and brain structure
Hypothesis 1a was investigated using the PROCESS model 1 (see Figure 1a). This model
was chosen to examine the moderating role of HIV-status to relationships between CCSE and
cognition. Hypothesis 1b was also investigated using PROCESS model 1 (see Figure 1b).
25
Relationships between CCSE and 8 volumetric ROIs were tested using HIV-status as A
moderator variable. ROIs included PFC total volume, PFC regional volumes (dlPFC, vlPFC,
OFC, anterior cingulate), amygdala, hippocampal, and insula volumes. FDR correction was
applied to individual PFC regions to account for multiple comparisons. Additionally, exploratory
analyses were run for all individual brain regions included in the PFC composite to clarify
specificity of any findings (e.g., rostral anterior cingulate, caudal anterior cingulate, lateral OFC,
medial OFC, superior frontal gyrus, caudal middle frontal gyrus, rostral middle frontal gyrus,
pars opercularis, pars triangularis, pars orbitalis; Table 6). Due to previous findings of laterality
reported in the literature, additional analyses were also run for left and right volumes separately
that are provided in Supplementary Table 3.
Aim 2. Examine if brain volume mediates relationships between CCSE and cognition
To evaluate if brain volume mediates relationships between CCSE and cognition,
mediation analyses were run using PROCESS model 4 (see Figure 1c). Given the lack of HIV
status interactions in aim 1, moderated mediation models were not utilized, and HIV status was
included as a covariate. CCSE served as the predictor (X), global cognition or individual
cognitive domains served as the respective dependent variables (Y), and brain volume served as
the mediator (M). 5000 bootstrap samples were generated with corresponding 95% confidence
intervals (CI).
Aim 3. Examine if CCSE is associated with cognitive change
For the cognitive change variable, z-scores were generated for global cognition and
individual cognitive times at baseline and at follow-up as described above. For global cognition
and each individual domain, follow-up z-scores were regressed on baseline z-scores and
unstandardized residuals were used as the dependent variable. Next, the unstandardized residual
26
was regressed on CCSE covarying for variables used in prior analyses and time-to-follow up
(years) (See Figure 1d). To interpret findings, CCSE groups were calculated based on standard
deviations: low CCSE = scores less than 1 standard deviation below mean, average CCSE =
scores between one standard deviation below and above mean, high CCSE = scores greater than
1 standard deviation above mean. Given concerns about sample sizes within low and high groups
(n = 11 each), CCSE groups were also created based on tertile cut-offs to aid interpretation.
Then, paired t-tests were used to compare baseline and follow-up z-scores within each group to
assess change.
Aim 4. Examine relationships between CCSE, vascular age discrepancy, individual vascular
risk factors, and cognition
Spearman correlations and logistic regression were used to test relationships between
CCSE and vascular risk variables (e.g., vascular age discrepancy, individual VRFs).
To assess relationships between vascular risk and cognition, the PROCESS macro (model
1) was used (Hayes, 2013). For vascular risk, we examined vascular age discrepancy. To reduce
separate analyses for PWLH and HIV- groups, HIV status and an interaction term (i.e., vascular
age discrepancy x HIV status) was included in every model. If the p-value for the interaction
term was less than 0.1, separate analyses were conducted for PLWH and HIV- groups.
Aim 5. Examine if vascular age discrepancy mediates or moderates relationships between
CCSE and cognition
If there was evidence for to test for a mediating relationship based on the above analyses
(i.e., CCSE is positively associated with vascular age discrepancy), PROCESS mediation will be
used to examine whether vascular age discrepancy mediates relationships between CCSE and
cognition. If not, relationships between CCSE and cognition will be repeated (Aim 1a) adding
27
vascular risk variables to the model to see if vascular risk attenuates relationships between CCSE
and cognition. Models will test both addition of vascular age discrepancy and addition of all
individual vascular risk factors (e.g., hypertension, T2DM, dyslipidemia, current smoking),
separately.
For moderation analyses, PROCESS moderation was used to examine the presence of
moderating relationships between CCSE and vascular age discrepancy on global cognition and
individual cognitive domains. Both 2-way interactions (vascular risk x CCSE) and 3-way
interactions (vascular risk x CCSE x HIV status) were examined.
Ancillary Analyses
After running primary analyses, additional analyses were run in order to clarify results
from study hypotheses. These analyses are referenced in the discussion.
Analysis 1. Examine if health-related stress masks HIV status interactions on cognition and
brain volume.
We repeated Aim 1 analyses without the inclusion of health-related stress covariate to
evaluate whether this covariate was suppressing any HIV status interactions.
Analysis 2. Examine if alternative lifetime stress indices are associated with cognition and
brain volume.
We repeated Aim 1 analyses with different cumulative stress exposure STRAIN indices
including total count and severity regardless of stressor type (e.g., acute stress, chronic stress),
acute stress count, acute stress severity, and chronic stress count. This was done to test whether
CCSE was more strongly related to cognition compared to other indices that did not include
subjective appraisal of exposure severity and/or chronic stressors.
Analysis 3. Examine if childhood CCSE is associated with limbic brain structures.
28
Given previous reports of relationships between early life stress exposure and subcortical
volumes in the amygdala and hippocampus using different stress measures, relationships
between early life CCSE and subcortical volumes were investigated.
Analysis 4. Examine relationships between individual vascular risk factors and cognition.
We conducted these additional analyses to clarify drivers of relationships between
vascular risk and cognition.
RESULTS
Descriptive Statistics
Demographic, socioeconomic, and psychosocial variables are shown for the whole
sample and PLWH and HIV- separately (Table 2). Relative to HIV- adults, PLWH had lower
total household income and income per family member (p’s < 0.001), greater current depressive
symptoms (p = 0.010) and CCSE (p = 0.035), and higher rates of previous depression (p =
0.001), past mania/psychosis (p = 0.032), past substance dependence (p < 0.001), and Hepatitis C
infection (p = 0.013). HIV status groups did not differ in age, sex, race, years of education,
premorbid estimated intelligence, current substance abuse (alcohol or marijuana only), or SES
(Hollingshead).
Regarding vascular risk factors, PLWH had higher rates of dyslipidemia than the HIV-
group (c
2
= 9.433, p = 0.002), and marginally higher 10-year CVD risk (Mann-Whitney U =
2616.00, p = 0.073). There were no other HIV status differences in vascular age discrepancy or
vascular risk factors (Table 3).
MRI Subset
For subset who completed MRI, PLWH had lower total household income and income
per family member (p’s < 0.001), higher health-related stress (p = 0.017), and higher rate of past
29
substance dependence (p < 0.001) (Table 4). HIV status groups did not differ in age, sex, race,
years of education, premorbid estimated intelligence, current substance abuse (alcohol or
marijuana), Hepatitis C infection, CCSE, or SES (Hollingshead). Comparison of those who
completed MRI versus not showed those who completed MRI had greater CCSE (p = 0.007) and
current alcohol or marijuana abuse (p = 0.049) (Supplemental Table 1). There were no
differences in any other demographic, SES, psychiatric, or substance use measures (p’s > 0.1).
Longitudinal Subset
For subset who completed follow-up, PLWH had lower total household income (p <
0.001), income per family member (p = 0.001), more current depressive symptoms (p = 0.024),
and higher rate of past substance dependence (p = 0.001) (Table 5). Comparison of those who
completed follow-up versus not showed the follow-up group had more PLWH (p = 0.034) and
higher rates of hepatitis C (p = 0.045) (Supplemental Table 2). There were no differences in any
other demographic, SES, psychiatric, or substance use measures (p’s > 0.1).
Correlates of CCSE
Correlation analyses demonstrated that CCSE was positively associated with recent
depressive symptoms (r = 0.238, p = 0.002), premorbid estimated intelligence (r = 0.180, p =
0.022), total household income (r = -0.175, p = 0.031), past mania/psychosis (p = 0.030), past
substance dependence (p = 0.015), past major depression (p = 0.006), and marginally associated
with current substance abuse (p = 0.074). There were no relationships with education (p =
0.599), SES (p = 0.241), age (p = 0.650), race (p = 0.442), or current major depression (p =
0.239).
Further, to determine relationships between childhood and adult stressor exposure, we
examined correlation between childhood CCSE and adulthood CCSE. Analyses showed
30
childhood CCSE was positively associated with adulthood CCSE (r = 0.485, p < 0.001), but did
not completely overlap indicating that these were distinct constructs.
Aim 1: Hypothesis 1a: CCSE will be negatively associated with global cognition and
processing speed, executive functioning, learning, and memory.
CCSE was negatively associated with global cognition (b = -0.08, SE = 0.03, t = -2.381,
p = 0.019, 95% CI: -0.14, -0.01), processing speed (b = -0.13, SE = 0.04, t = =3.014, p = 0.003,
95% CI: -0.22, -0.05), and executive functioning (b = -0.07, SE = 0.03, t = -2.088, p = 0.039,
95% CI: -0.13, -0.00). There were no relationships with learning (p = 0.240) or memory (p =
0.317). Significant findings were unchanged after excluding participants who met criteria for
current major depression based upon the SCID (n = 4), or past mania or psychosis (n = 6), or
current alcohol or marijuana abuse (n = 13). When excluding all groups (n = 139), significant
findings remained for global cognition and processing speed, however executive functioning was
attenuated (p < 0.1).
There were no statistically significant HIV group differences in processing speed (p =
0.127), executive functioning (p = 0.350), learning (p = 0.792), or memory (p = 0.815).
There were no statistically significant CCSE × HIV status interactions on global
cognition (p = 0.209) or any individual domains (processing speed: p = 0.211; executive
functioning: p = 0.877; learning: p = 0.287; memory: p = 0.228).
Aim 1: Hypothesis 1b: CCSE will be negatively associated with volumes in the PFC,
hippocampus, and insula, and positively associated with amygdala volume.
CCSE was negatively associated with PFC volume (b = -16.33, SE = 5.93, t = -2.751, p =
0.007, 95% CI: -28.16, -4.51), including the dlPFC (b = -33.28, SE = 14.63, t = -2.275, p =
0.026, 95% CI: -62.42, -4.14), and OFC (b = -14.49, SE = 5.32, t = -2.727, p = 0.008, 95% CI: -
31
25.08, -3.91). CCSE was marginally negatively associated with vlPFC volume (p = 0.097). There
were no relationships with anterior cingulate volume (p = 0.124), amygdala volume (p = 0.563),
hippocampal volume (p = 0.717), or insula volume (p = 0.435). The PFC finding remained
significant after additionally excluding for current major depression, current alcohol or marijuana
substance abuse, and past mania or psychosis (n = 72).
Additional analyses were run to examine relationships with individual ROIs within the
PFC (Table 6). Briefly, CCSE was inversely associated with volumes in the medial OFC (t = -
2.457, p = 0.016), and nonsignificant trending relationships in the superior frontal gyrus (t = -
1.973, p = 0.052) and lateral OFC (t = -1.878, p = 0.064). Relationships were not significant in
the rostral anterior cingulate (p = 0.336), caudal anterior cingulate (p = 0.101), caudal middle
frontal gyrus (p = 0.199), rostral middle frontal gyrus (p = 0.123), pars opercularis (p = 0.220),
pars triangularis (p = 0.117), and pars orbitalis (p = 0.679). Given evidence of laterality in
previous studies, additional analyses were conducted in ROIs in each hemisphere (see
Supplemental Table 3).
PLWH had significantly lower volumes in the PFC composite (b = -353.86, SE = 168.61,
t = -2.099, p = 0.039, 95% CI: -689.67, -18.05) and OFC (b = -316.38, SE = 150.97, t = -2.096, p
= 0.039, 95% CI: -617.06, -15.70), and marginally smaller volumes in the vlPFC (p = 0.096).
HIV status groups did not differ on volumes in the dlPFC (p = 0.119), anterior cingulate (p =
0.170), amygdala (p = 0.376), hippocampus (p = 0.212), or insula (p = 0.814).
There was no interaction between CCSE and HIV status on PFC volume (p = 0.180),
anterior cingulate volume (p = 0.656), dlPFC volume (p = 0.248), vlPFC volume (p = 0.321),
OFC volume (p = 0.241), amygdala volume (p = 0.932), hippocampal volume (p = 0.490), or
insula volume (p = 0.814).
32
Aim 2: Hypothesis 2: Brain volume will mediate significant relationships between CCSE
and cognition.
Based on findings from aims 1 and 2, path analyses were conducted to examine whether
brain volumes mediate relationships between CCSE and outcomes of global cognition,
processing speed, and executive functioning.
Global Cognition
CCSE was significantly associated with PFC volume (a: b = -12.77, SE = 5.35, t = -
2.387, p = 0.020). When controlling for CCSE, PFC volume was not significantly associated
with global cognition (b: b = 0.00, SE = 0.00, t = 1.479, p = 0.143). The estimated total effect of
CCSE on global cognition was significant (c: b = -0.10, SE = 0.03, t = -2.779, p = 0.007, 95%
CI: -0.17, -0.03) and the estimated direct effect of CCSE was significant (c’: b = -0.08, SE =
0.04, t = -2.314, p = 0.023). The indirect effect of CCSE on global cognition through PFC
volume was not significant (95% Bootstrap CI: -0.04, 0.00) thus there was no evidence of partial
or full mediation. Follow-up analyses showed no evidence of moderated mediation by HIV status
(95% Bootstrap CI: -0.01, 0.03).
Processing Speed
CCSE was significantly associated with PFC volume (a: b = -12.77, SE = 5.35, t = -
2.387, p = 0.020) (Figure 8). When controlling for CCSE, PFC volume was significantly
associated with processing speed (b: b = 0.00, SE = 0.00, t = 2.213, p = 0.030). The estimated
total effect of CCSE on processing speed was significant (c: b = -0.12, SE = 0.05, t = -2.341, p =
0.022, 95% CI: -0.22, -0.02) while the estimated direct effect of CCSE was not significant (c’: b
= -0.09, SE = 0.05, t = -1.735, p = 0.087). The indirect effect of CCSE on processing speed
through PFC volume was significant (95% Bootstrap CI: : -0.08, -0.00). Given that the direct
33
path from CCSE to processing speed (c’) was not significant, the relationship between CCSE
and processing speed was fully mediated by PFC volume. Follow-up analyses showed no
evidence of moderated mediation by HIV status (95% Bootstrap CI: -0.01, 0.06).
Additional analyses demonstrated that the relationship between CCSE and processing
speed was not mediated by OFC volume (95% Bootstrap CI: -0.09, 0.00) or dlPFC volume (95%
Bootstrap CI: -0.06, 0.00).
Executive Functioning
CCSE was significantly associated with PFC volume (a: b = -12.80, SE = 5.37, t = -
2.385, p = 0.019). When controlling for CCSE, PFC volume was not associated with executive
functioning (b: b = 0.00, SE = 0.00, t = 1.276, p = 0.206). The estimated total effect of CCSE
was significant (c: b = -0.09, SE = 0.03, t = -2.586, p = 0.012), and the direct effect was
significant (c’: b = -0.08, SE = 0.03, t = -2.168, p = 0.033). The indirect effect of CCSE on
executive functioning through PFC volume was not significant (95% Bootstrap CI: -0.04, 0.01)
suggesting there was no partial or full mediation present. Follow-up analyses showed no
evidence of moderated mediation by HIV status (95% Bootstrap CI: -0.01, 0.03).
The relationship was also not mediated by OFC volume (95% Bootstrap CI: -0.04, 0.00)
or dlPFC volume (95% Bootstrap CI: -0.03, 0.01).
Aim 3: Hypothesis 3: Increased CCSE will be associated with cognitive decline from
baseline to 2-year follow-up.
CCSE was significantly associated with change in learning (b = 0.01, SE = 0.01, t =
2.664, p = 0.010, 95% CI: 0.00, 0.02). Follow-up analyses showed that the high CSSE group (n =
11) showed increased performance in learning over time (D = 0.30 SD, t = 2.390, p = 0.038),
while the other groups did not change (low CSSE: D = -0.22 SD, t = 1.600, p = 0.141; average
34
CSSE: D = -0.02 SD, t = -0.186, p = 0.853). Alternatively, when CCSE was split into tertile
groups, no group demonstrated significant changes (lowest tertile: D = -0.17 SD, t = -1.543, p =
0.135; middle tertile: D = 0.04 SD, t = 0.288, p = 0.776; highest tertile: D = 0.13 SD, t = 1.268, p
= 0.216).
CCSE was not associated with change in global cognition scores (p = 0.493), processing
speed (p = 0.798), executive functioning (p = 0.135), or memory (p = 0.122). There were no
significant effects for HIV status or CCSE-by-HIV-status (p’s > 0.1).
Aim 4: Hypothesis 4a: Increased CCSE will be associated with increased vascular risk and
increased rates of vascular risk factors.
CCSE was not associated with vascular age discrepancy (p = 0.495), or presence of
vascular risk factors (i.e., presence of 1 or more risk factors) (p = 0.265). Regarding individual
risk factors, CCSE was marginally negatively associated with current smoking such that
increased CCSE was associated with lower likelihood of current smoking (exp(B) = 0.974, p =
0.057). CCSE was not related to dyslipidemia (p = 0.591), obesity (p = 0.424), T2DM (p =
0.705), or hypertension (p = 0.320). There were no HIV status interactions (p’s > 0.1).
Aim 4: Hypothesis 4b: Vascular risk will be negatively associated with cognition such that
increased vascular risk is associated with worse cognitive performance.
Vascular age discrepancy was not associated with any cognitive domains: global
cognition (p = 0.863), processing speed (p = 0.560), executive functioning (p = 0.654), learning
(p = 0.549), or memory (p = 0.500). There was no evidence of any HIV status or interactive
effects (p’s > 0.1).
Aim 5: Hypothesis 5a Vascular age discrepancy will partially mediate relationships
between stress and cognition.
35
Given null findings assessing relationships between CCSE and vascular risk, no
mediation analyses were performed. However, analyses assessing relationships between CCSE
and cognition were performed (replication of Aim 1a) with the addition of vascular risk (e.g.,
vascular age discrepancy, individual vascular risk factors) to the model to see if vascular risk
attenuated findings.
After adding vascular age discrepancy to the model, findings remained significant for
global cognition (b = -0.07, SE = 0.03, t = -2.168, p = 0.034, 95% CI: -0.14, -0.01), and
processing speed (b = -0.13, SE = 0.05, t = -2.789, p = 0.006, 95% CI: -0.22, -0.04). Results
were slightly attenuated for executive functioning (p = 0.066). Results remained nonsignificant
for learning (p = 0.267) and memory (p = 0.325). There were no main effects for vascular age
discrepancy on global cognition (p = 0.759) or any individual domain: processing speed (p =
0.757), executive functioning (p = 0.199), learning (p = 0.780), or memory (p = 0.644).
Results were similar when controlling for all individual vascular risk factors within the
model (e.g., hypertension, dyslipidemia, obesity, T2DM, and current smoking). Findings
remained significant for global cognition (b - =0.09, SE = 0.03, t = -2.745, p = 0.007, 95% CI: -
0.16, -0.03), processing speed (b = -0.15, SE = 0.05, t = -3.197, p = 0.002, 95% CI: -0.24, -0.06),
and executive functioning (b = -0.08, SE = 0.04, t = -2.123, p = 0.035, 95% CI: -0.15, -0.01).
Results remained nonsignificant for learning (p = 0.216) and memory (p = 0.155).
The were no main effects for HIV status or HIV status interactions in any model (p’s >
0.1).
Aim 5: Hypothesis 5b: Vascular risk and CCSE will have an interactive effect on cognitive
outcomes such that those with high vascular risk and high CCSE will have lowest scores on
cognitive outcomes.
36
Vascular Age Discrepancy
There were no significant 2-way interactions (CCSE x vascular age discrepancy) on
global cognition (p = 0.285), processing speed (p = 0.456), executive functioning (p = 0.336),
learning (p = 0.381), or memory (p = 0.731). Similarly, there were no 3-way interactions (CCSE,
x vascular age discrepancy, and HIV status on global cognition (p = 0.739), processing speed (p
= 0.300), executive functioning (p = 0.458), learning (p = 0.306), and memory (p = 0.276).
Ancillary Analyses
To better understand the findings from Aims 1-5, a series of ancillary analyses were
conducted to answer specific questions. These questions and analyses are described below:
Analysis 1. Did the inclusion of a covariate related to health-related stress mask HIV status or
interactive effects?
When health-related stress was removed as a covariate, there were still no significant
interactions with HIV status on global cognition, individual cognitive domains, or volumes in the
PFC, amygdala, hippocampus, and insula (p’s > 0.1).
Analysis 2. Are other indices of lifetime stress also related to cognition and PFC volume in
this sample?
Examination of relationships between other STRAIN stressor indices and cognition and
PFC volume showed our chosen variable, CCSE (i.e., chronic stressor severity), was most
consistently related to outcomes of interest and most strongly related (see Supplemental Table 4).
Total stressor count was negatively associated with processing speed (p = 0.045). Total stressor
severity was negatively associated with global cognition (p = 0.025), processing speed (p =
0.006), and PFC volume (p = 0.048). Chronic stressor count was negatively associated with
processing speed (p = 0.047) and PFC volume (p = 0.011). Acute stressor severity was
37
negatively associated with processing speed (p = 0.041). No other relationships were significant
(p’s > 0.1).
Analysis 3. Is chronic stress in early life related to brain volumes of interest?
Childhood CCSE was not associated with volumes in the amygdala (p = 0.192),
hippocampus (p = 0.868), or insula (p = 0.837). Findings remained non-significant once
removing health-related stressor exposure covariate and separating into left and right
hemispheres (p’s > 0.1, data not shown).
Analysis 4. Are individual vascular risk factors associated with cognition?
Participants with dyslipidemia had lower scores on memory (F1,151= 4.762, p = 0.031,
partial h
2
= 0.031). Participants with obesity had marginally greater scores on executive
functioning (F1,150 = 3.518, p = 0.063, partial h
2
= 0.023). There were no other main effects for
individual vascular risk factors (p’s > 0.1).
There was a significant interaction between HIV status and current smoking on executive
functioning (F1,150 = 8.836, p = 0.003, partial h
2
= 0.056), such that current smoking was
associated with better executive functioning in PLWH (F1,57 = 6.785, p = 0.012, partial h
2
=
0.106), but was not associated with executive functioning in the HIV- group (p = 0.469). There
were no other significant interactions (p’s > 0.1).
DISCUSSION
To our knowledge, this is the first study to report relationships between lifetime stressor
exposure and cognition and brain structure in an adult sample. The current study provides
important insights into the relationships between cumulative chronic stress exposure and brain
structure and functioning in a diverse group of adults with high risk for stress exposure (i.e.,
PLWH). Findings suggest cumulative stressor exposure over the lifetime, especially chronic
38
stress exposure, may negatively affect cognitive functioning and brain structures necessary to
both cognition and emotion regulation.
Aim 1a – Is CCSE Related to Cognition?
This is the first study to use the STRAIN to understand relationships between lifetime
cumulative chronic stressor exposure over the lifetime and brain function among PLWH. We
found cumulative chronic stressor exposure to be inversely associated with several cognitive
abilities, including processing speed, and executive functioning. The results lend support to the
theory that the effects of stressor exposure are additive and exert a cumulative effect on
biological processes that promote increase risk for disease and negative health outcomes
(Graham et al., 2006; Lupien et al., 2009). Recent STRAIN validation studies have reported
inconsistent findings in the relationship between total lifetime stressor exposure and executive
functioning (Cazassa et al., 2020; Slavich & Shields, 2018). Compared to these prior studies, our
study benefitted from a comprehensive neuropsychological battery that uses at least two tests to
define each cognitive domain. While there are no known studies of lifetime stress and cognition,
these results are consistent with another study that found lifetime stress as measured by the
Youth Life Stress Interview was associated with spatial working memory in adolescents whereas
recent stress (e.g., in last year) was not (Hanson et al., 2012).
Processing speed was most strongly associated with CCSE in our study. The relationship
between stress and processing speed has been found in other studies (Caswell et al., 2003; Clark
et al., 2012; Spies et al., 2017). One study showed processing speed mediated the relationship
between psychosocial stress and global cognition in older adults suggesting stress-related deficits
in processing speed may be responsible for relationships with overall cognitive dysfunction
(Foong et al., 2018) . Also consistent with past literature examining recent perceived stress,
39
CCSE was associated with executive functioning (Ohman et al., 2007; Rubin et al., 2015). We
did not find relationships with learning or memory, which have been reported in prior studies
(Ohman et al., 2007; Rubin et al., 2015). Notably, our learning and memory constructs differed
from some prior studies that only used verbal learning/memory measures.
Contrary to expectations, we did not find any significant differences between HIV status
groups on cognition, despite evidence that asymptomatic neurocognitive impairment and mild
neurocognitive impairment persists in the cART era (Heaton et al., 2010). However, the
Women’s Interagency HIV Study reported the effect size for HIV status on cognitive functioning
was very small (e.g., 0.05-0.09 standard deviation units) and was smaller than that of other
correlates of cognitive function (e.g., education, age, race, income, reading level) (Maki et al.,
2015). It is likely that our lack of cognitive differences among HIV status groups is due to our
well-matched HIV- comparison group, and inclusion of important covariates (e.g., premorbid
estimated intelligence, race) in cognitive analyses that are not universally included in studies.
Our groups did not differ on key demographic variables including age, sex, race, education, and
estimated premorbid intelligence. Groups did differ on household income, which is likely reflects
that the PLWH group were more likely to report disability income. Further, our inclusion of
covariates such as race, premorbid estimated intelligence, and depressive symptoms may have
attenuated any group differences on cognitive measures. Our group recently published a
systematic review of vascular risk factors and cognition in PLWH, which revealed that many
studies do not include race or SES in models (McIntosh et al., 2021).
Another possible explanation for lack of HIV status differences in cognition may be that
our HIV cohort is healthier than others reported in the literature. Our HIV cohort reported high
cART use (>98%), and the mean current CD4 count was 732 with the majority of participants
40
(73.4%) with normal CD4 count (e.g., >500). While a sizeable amount had nadir count <200
indicative of AIDS diagnosis (43.3%), the majority of our participants were not recently infected
(mean duration 21 years) thus there may also be a survivor effect present.
While HIV status did not influence many of our demographic variables, PLWH did have
greater current depressive symptoms, CCSE, and health-related CCSE relative to the HIV-
group. The pattern of differences in depressive symptoms and stressor exposure is consistent
with the extant literature showing PLWH have higher rates of trauma, stress exposure, and
psychiatric disorders relative to HIV-uninfected population (Leserman et al., 2008; Machtinger
et al., 2012; Pence et al., 2007).
Contrary to hypotheses, CCSE-cognitive relationships were not moderated by HIV status.
The null interaction between HIV status and lifetime CCSE may be explained by decreased
power (e.g., smaller sample size) compared to other studies that have reported interactions
between HIV-status and stressor variables on cognition (Pukay-Martin et al., 2003; Rubin et al.,
2015; Spies et al., 2016). Furthermore, unlike prior studies in PLWH, this sample had both male
and female participants which may have added variance to our models. The extant literature
shows that men and women may have different reactions to stressful life experiences or be
exposed to different stressors based on traditional gender roles (Cohen et al., 2019). Previous
studies have documented sex differences in cortisol responses to psychological stressors and
relationships between cortisol production and cognitive performance (Reschke-Hernández et al.,
2017; Schoofs et al., 2013). Given the sample is predominantly male, we were underpowered to
perform separate analyses for males and females, but could be explored with additional data
collection.
41
Further, it is noteworthy that relationships between CCSE and cognition were
independent of current depressive symptoms and remained significant after removing
participants who met criteria for current major depression. This is important as depression is well
known to negatively affect cognition, including attention, executive functioning, and memory
(Rock et al., 2014). Depression and other psychiatric disorders can be caused (e.g., diathesis-
stress model) or exacerbated by stress exposures (Colodro-Conde et al., 2018; Hammen et al.,
2000). Accounting for depressive symptoms in examinations of chronic stress and cognition may
be particularly important as chronic stress is a greater predictor of depressive symptoms relative
to acute stress (McGonagle & Kessler, 1990). In a recent study utilizing the STRAIN,
cumulative stress over the course of one’s lifetime was associated with increased depression and
anxiety symptoms (McLoughlin et al., 2021). Thus, considering relationships between stress,
depression and cognition, it is important to account for depression when isolating the cognitive
correlates of stress. One study found that after controlling for depressive symptoms relationships
between perceived racial discrimination and cognition were attenuated (Barnes et al., 2012).
However, the majority of studies that reported relationships between early life stress or recent
perceived stress and cognition found these relationships were independent of current psychiatric
symptoms, including depression (Aggarwal, Wilson, et al., 2014; Clark et al., 2012; Pukay-
Martin et al., 2003; Rubin et al., 2015).
Aim 1b – Is CCSE Related to Brain Structures Implicated in Stress and Emotion
Regulation?
There is a wealth of literature that supports increased stress is related to functional and
structural changes in the brain that regulate the stress response and emotion regulation including
the PFC, hippocampus, and amygdala (Hanson et al., 2015; McEwen et al., 2016). Structures
42
related to HPA axis function are integral in adaptation to stressors (allostasis), and its
dysfunction is associated with maladaptive responses to stressors (excessive allostatic load)
(McEwen & Stellar, 1993).
In the current study, we found CCSE was associated with PFC volume in the expected
direction (i.e., higher was CCSE associated with reduced PFC volume). The relationship between
CCSE and PFC volume adds to the extant literature documenting relationships between stress
and PFC volume in human adults (Ansell et al., 2012; Gianaros et al., 2007; Moreno et al., 2017;
L. H. Rubin et al., 2016). We found CCSE was associated with reduced volumes in several
regions of the PFC. The relationship between stress and OFC volume has been reported
previously (Ansell et al., 2012; Gianaros et al., 2007). Further, our finding in the medial OFC is
consistent with prior research that chronic stress and corticosteroid-induced stress is associated
with dendritic shortening and atrophy in the medial PFC (McEwen, 2007). Similarly, a
neuroimaging study showed higher perceived stress in women living with HIV was associated
with greater fMRI BOLD deactivation in the PFC during a memory recognition task (L.H. Rubin
et al., 2016). Our findings in the dlPFC and aspects of the vlPFC are also consistent with prior
studies (Moreno et al., 2017), and suggest that the effect of stress on the brain may not be limited
to the medial PFC.
There were no findings in subcortical regions related to the HPA axis and limbic system
such as the hippocampus, amygdala, and insula. Despite strong theoretical underpinnings
supporting a relationship between stress and hippocampal volume, there are mixed findings in
the literature. Inverse relationships between perceived stress and hippocampal volume have been
reported in adolescents and older adults (Gianaros et al., 2007; Piccolo & Noble, 2018;
Zimmerman et al., 2016). In contrast, several studies have not reported such a relationship
43
(Ansell et al., 2012; Moreno et al., 2017; L. H. Rubin et al., 2016; Wu et al., 2021). Interestingly,
these aforementioned studies all found relationships between stress and PFC volumes,
demonstrating that these studies did find expected associations between stress and other brain
structures. One possibility for conflicting findings in the hippocampus may be the brain may be
more sensitive to stress-related damage to the hippocampus at certain ages. For example, the
impact of stress exposure may be particularly detrimental in early age or late age due to
developmental changes (e.g., rates of growth or atrophy). In support of this, a recent study
showed relationships between perceived stress and structural brain volume in adolescents and
middle-aged adults had opposite relationships (e.g. stress negatively associated with OFC in
adults but positively associated in adolescents) (Wu et al., 2021). Similarly, some animal models
have shown chronic psychosocial stress is more strongly tied to decreased dentate gyrus cell
proliferation in older compared to younger animals (Simon et al., 2005).
There was no relationship between CCSE and amygdala volumes. While the amygdala is
hypothesized to be vulnerable to the effects of stress, the majority of studies that have
documented positive relationships between amygdala volume and stress have been limited to
early life stress (Clark et al., 2012; Mehta et al., 2009; Tottenham et al., 2010). It is hypothesized
that early life stress may be particularly detrimental to amygdala function and structure due to
early development of this brain region in childhood (Tottenham & Sheridan, 2009). In the recent
stress literature, mostly compromised of studies using the perceived stress scale, several studies
have not found relationships between recent stress and amygdala volume (Ansell et al., 2012;
Gianaros et al., 2007; Moreno et al., 2017), or have found recent stress was associated with lower
amygdala volume (Sublette et al., 2016; Wu et al., 2021). The one neuroimaging study that did
measure cumulative adverse life events, which most closely resembles our measure, also did not
44
find relationships with amygdala volume (Ansell et al., 2012). In sum, the relationship between
stress and amygdala volume outside of early life stress is mixed, and future studies are warranted
to understand how types and timing of stress may affect these relationships.
Similarly, we did not find associations between stressor exposure and the insula. The
insula, while not as prominently featured in literature regarding stress and brain structure, is a
paralimbic structure that is believed to be important for emotional experience (Uddin et al.,
2017). One study showed cumulative adversity was inversely associated with volumes in the
insular cortex, as well as the medial PFC and anterior cingulate (Ansell et al., 2012). Similarly,
another study showed perceived stress in middle-aged adults was negatively associated with
insula volume (Wu et al., 2021). Notably, these studies samples were different from our sample
in several key ways; they were younger than our sample and did not include PLWH. Differences
in study demographics and stressor measures thus make it difficult to compare our findings.
One explanation for the general lack of findings in subcortical regions and the insula is
small sample size. The sample size of the current study was lower than several of the studies
cited above, and also contained significant heterogeneity as it included both PLWH and HIV-
adults, and men and women. Further, there was also heterogeneity in the age of our participants
(range 32-73 years). Given prior research that brain regions may be more susceptible to stress
exposure depending on age and development, this may have obscured findings.
Aim 2 – Does Brain Volume Mediate Relationships between CCSE and Cognition?
Given the literature on relationships between stress, cognition and brain volumes, it is
prudent to explore whether associations with stressor exposure and cognition are at least in part
due to volume in brain regions sensitive to stress. In the present study, mediation analysis
showed the relationship between CCSE and processing speed was mediated by PFC volume.
45
Though we cannot infer causality, the results of the mediation model may suggest that lifetime
stressor-related deficits in processing speed are driven by structural abnormalities in the PFC.
Processing speed is associated with both white matter integrity and gray matter volume,
including PFC volumes (Eckert et al., 2010; Kennedy & Raz, 2009). Processing speed does not
involve a single neural system, but involves several neural networks. Consistent with the tests in
our battery, processing speed tasks can encompass multiple skills including visuoperception
skills (e.g., stimulus detection, stimulus discrimination, visual tracking), motor skills, and verbal
skills (e.g., reading and language abilities), thus processing speed can be associated with several
different brain regions depending on the specific task, which may explain why our significant
findings were observed using the PFC composite as opposed to parcellated regions. Future
studies may benefit from using a voxel-based approach to better understand the relationship
between processing speed and PFC structure in the context of stress.
While CCSE was associated with executive functioning, this relationship was not
mediated by PFC volume. The definition of executive functions is very broad and includes
myriad cognitive skills including set-shifting, working memory, phonemic fluency, problem
solving, inhibitory control, and decision-making (Baddeley & Hitch, 1974; Roberts Jr &
Pennington, 1996). Though executive functions are thought to be largely governed by the PFC,
individual aspects of executive functioning correspond to different areas of the PFC. For
example, Trails B has been linked to dlPFC structure and function in numerous studies (Shaked
et al., 2018). Consistent with the broad definition of executive functions, our executive
functioning composite included tasks that assess its various abilities including working memory
and set-shifting/mental flexibility (Trails B), inhibitory control (Stroop Interference), phonemic
fluency (FAS), and working memory (Letter Number Sequencing). Further, some tasks were
46
timed (e.g., Trails B, FAS) while others were not (Letter-Number Sequencing) or accounted for
processing speed (Stroop Interference). Thus, the heterogeneity in both executive function tasks
may have diluted the association with PFC volume and any mediating relationships.
Few studies have used proper mediation techniques to investigate this question. One
study in adolescents showed lifetime stress exposure was inversely associated with spatial
working memory performance, which was mediated by PFC volumes (Hanson et al., 2012). In a
study of SES, which is closely related to stressor exposure, dlPFC volume was found to partially
mediate the relationship between SES and executive function (Shaked et al., 2018). Future
research would benefit from similar mediation analyses to understand if relationships between
stress and cognition are mediated by structural alterations in the brain.
Aim 3 – Does CCSE Predict Cognitive Decline?
Our results showed there was non-significant change in global cognition and most
domains assessed. However, contrary to hypotheses, we found those with high CCSE (above one
standard deviation) showed increased learning scores at follow-up. However, CCSE tertile
groups did not differ suggesting that findings may be driven by those at extreme ends of
spectrum only. Post-hoc analyses showed that both the middle and highest tertile tended to
improve, albeit nonsignificantly, over time, which is likely consistent with practice effects over a
shorter follow-up time (Dikmen et al., 1999). Further, those with lower CCSE had lower
estimated premorbid intelligence and demonstrated stable performances. The pattern of lower
estimated premorbid intelligence among lower CCSE groups may indicate they have lower
cognitive reserve and less likely to benefit from practice effects (Tucker & Stern, 2011). One
study reported a similar finding where chronic stressors in older persons, such as illness of a
partner or relative and interpersonal conflict, were associated with better cognitive function over
47
time (Comijs et al., 2011). However, for the aforementioned reasons, we are cautious in
interpreting our finding. It is also important to note that the increase (0.3 SDs) is not clinically
meaningfully.
Though we hypothesized that greater CCSE would be associated with cognitive decline,
the majority of null findings is not entirely surprising given mixed findings in the literature and
heterogeneous samples (e.g., age) and methodologies (e.g., stress measurement, time to follow
up). Our null findings may also be in part explained by a relatively short length of time between
assessments, and relatively young age of participants. There is evidence that changes in
cognition are accelerated at higher ages (e.g., greater than 60 years-old) (Salthouse, 2009).
Further there is evidence that general cognitive decline characterized by declines in several
domains increases as we age (Tucker-Drob et al., 2019).
Aim 4a – Is CCSE associated with vascular risk?
Contrary to hypotheses, CCSE was not associated with vascular age discrepancy, or
individual risk factors. This is inconsistent with a previous study using the STRAIN that showed
cumulative life stress exposure was associated with greater metabolic risk using a cumulative
risk score that included waist circumference, glucose, cholesterol, triglycerides, and systolic
blood pressure in pre-menopausal mothers (Kurtzman et al., 2012). In another study of maternal
child caregivers, mothers of children with autistic spectrum disorder had worse metabolic health
as demonstrated by greater waist circumference, lower HDL cholesterol, increased rates of
insulin resistance, and higher triglycerides, and greater reward-based eating (Radin et al., 2019).
In a recent study of women living with HIV, psychosocial stress as defined by depressive
symptoms, perceived stress, and PTSD symptoms, was associated with increased carotid plaques
among women living with HIV but not HIV-uninfected women suggesting that women living
48
with HIV may be particularly vulnerable to stressor-associated vascular risk (Levy et al., 2020).
It is possible that our findings were influenced by our heavily male sample. Previous literature
shows evidence for gender differences in relationships between stress and vascular risk factors
and CVD outcomes (Taylor et al., 2018). Given gender differences in stressor exposure and
types of stressor exposure, this may partly account for null findings.
While considering our null findings, it is important to note that our sample, especially
PLWH, may differ in the types of stressors they are exposed to given their health status and
disability. For example, less than half of the sample reported being employed or self-employed,
and the majority of our PLWH group receive disability-based income (e.g., SSI). This is relevant
as several studies have shown job-related stress is associated with cardiovascular outcomes
(Belkic et al., 2004). A recent study showed that stress at work (e.g., effort-reward imbalance)
was associated with intima media thickness in late middle-aged adults (Mauss et al., 2020). In
adults with diabetes, higher work-related stress was related to greater hemoglobin A1c values,
while there was no significant relationship between frequency of daily stress and glycemic
control, highlighting the unique contribution of work-related stress to metabolic health (Walker
et al., 2020). While other studies have shown unemployment or inadequate employment is
associated with increased risk for CVD (Gallo et al., 2004; Hergenrather et al., 2015), this might
not be applicable to populations with chronic illness or disability.
The finding that CCSE was negatively associated with current smoking was surprising.
Chronic stress is hypothesized to increase risk for cardiovascular disease and cerebrovascular
disease in part through behavioral responses to stress including cigarette smoking. Further,
consistent with higher rates of trauma and stressful events in PLWH, cigarette smoking is very
prevalent in PLWH (Mdodo et al., 2015). Cigarette smoking increases risk for vascular disease
49
in part through atherosclerosis which causes blood pressure to increase; further cigarette smoking
and hypertension have interactive effect on carotid intima media thickness (Liang et al., 2001).
Thus, the unexpected pattern for smoking might also affect relationships between stress and
hypertension in this sample.
Aim 4b – Is vascular age discrepancy associated with cognition?
Contrary to hypotheses, we also did not find a relationship between vascular age
discrepancy and cognition. Two prior studies have used the same algorithm by the Framingham
Heart Study to examine relationships between vascular risk and cognition in HIV (Chow et al.,
2020; Wright et al., 2015). One study reported 10-year CVD risk was negatively associated with
global cognition in a large sample of cART-naïve HIV-infected adults (Wright et al., 2015).
More recently, another study showed 10-year CVD risk predicted cognitive decline among
women living with HIV (Chow et al., 2020). As described in detail below, the lack of biological
data to better characterize vascular risk (e.g., lipid panel, glucose) may have weakened the
strength of our vascular age variable.
In addition to limitations regarding measurement of vascular risk described below,
largely null findings may be attributable to lower rates smoking and type 2 diabetes compared to
other HIV cohorts and exclusion of individuals with CVD. For example, only 21% of our PLWH
reported current smoking, while the CDC estimated current smoking prevalence to be 34% in
PLWH accessing medical care (Frazier et al., 2018). Smoking rates may be even higher in other
countries; for example, in a recent large study in Italy, they reported 52% of their PLWH sample
smoked compared to 26% of controls (De Socio et al., 2020). In a recent meta-analysis, our
group showed T2DM and CVD were strongest predictors of cognitive impairment (McIntosh et
50
al., 2021). Thus, our low prevalence of T2DM and exclusion of individuals with CVD may have
contributed to some of our null findings.
Aim 5. Does vascular age discrepancy mediate or moderate relationships between CCSE
and cognition?
Contrary to hypotheses, vascular risk did not explain relationships between stress and
cognition. However, our findings are consistent with previous studies assessing relationships
between recent perceived stress and cognition that showed vascular risk factors did not
significantly attenuate relationships (Aggarwal, Wilson, et al., 2014; Turner et al., 2017). As
discussed previously, there could be several reasons for lack of findings regarding relationship
between stress and vascular risk, and vascular risk and cognition. However, given the extensive
literature regarding stress and vascular risk and cardiovascular disease, and vascular risk and
cognition, this question should be continued to explored with larger samples that are adequately
powered to test moderating effects such as type of stressor, age and sex.
To our knowledge, few if any studies have examined interactions between stress and
vascular risk on cognition thus it is difficult to compare our findings with the literature. We
hypothesized that the relationship between stress and cognition would be strengthened in those
with vascular risk. This hypothesis was based on previous findings in the literature showing
common effects of stress and vascular risk on shared cognitive domains including processing
speed, executive functioning, and learning/memory. While our hypotheses were not confirmed in
the present sample, this is an important question that should continue to be explored in both
HIV-uninfected and HIV populations. Better understanding of relationships between stress and
vascular risk on the brain will help better characterize cognitive dysfunction and risk for
cognitive decline in PLWH.
51
Ancillary Analyses
Analysis 1 – Does health-related stress mask HIV status effects and interactions on cognition?
Elimination of health-related stress from cognitive models did not change null HIV status
interactions suggesting that this variable did not suppress interactions in original analyses (p’s >
0.1). Thus, lack of HIV status differences in interactions in primary analyses are likely due to
reasons hypothesized above (e.g., healthy sample, well-matched HIV- group).
Analysis 2 – Are other lifetime stress measures associated with cognition and PFC volume?
In examining the specificity of our stressor variable to reported findings, we found CCSE
was consistently more strongly related to cognition than other stressor indices. Further, CCSE
was the only stressor index significantly associated with executive functioning. With the
exception of one finding in processing speed, cumulative acute stressor indices were not
significantly associated with cognition or PFC volume. Regarding brain structure analyses, we
found that total exposure severity, including both acute and chronic stressors, and chronic
stressor count was also related to PFC volume, while all other variables were not including total
stressor count, total acute stressor count, and total acute severity. Results of these exploratory
analyses are in line with hypotheses, and bolster the theory that stress appraisal (e.g., severity of
exposure) influences the impact of stressor exposure on health and behavior. Further, chronic
stressor indices were more strongly related to outcomes than acute stressor indices. Again, these
findings are consistent with hypotheses and the extant literature that argue that chronic stressors
are more impactful due to their prolonged duration.
Analysis 3 – Is early life CCSE related to subcortical volumes?
Early life CCSE was not associated with volumes in the hippocampus, amygdala, or
insula. While these relationships have been documented in prior studies, this is the first study to
52
our knowledge to be using the STRAIN to capture early life CCSE and it may differ from
previous methodologies that have focused more on traumatic experiences. In addition, some
studies that have documented relationships between early life adversity and
amygdala/hippocampal brain volumes have been performed in very specific and highly stressed
populations, such as children raised in orphanages (Tottenham et al., 2010). Compared to a
previous study in HIV that found enlarged amygdale in participants with early life stress, our
sample was on average 10 years older and used different methodology to quantify stress (early
life stress questionnaire) (Clark et al., 2012).
Analysis 4 – Are individual vascular risk factors related to cognition?
Additional analyses using individual vascular risk factors revealed some expected and
unexpected findings. Consistent with what we would expect, dyslipidemia was associated with
worse memory performance. Relationships between dyslipidemia and cognition and cognitive
decline have been reported in several HIV cohorts (Ciccarelli et al., 2015; Gomez et al., 2017;
Mukerji et al., 2016; Wright et al., 2010). Importantly, dyslipidemia was not part of our vascular
age discrepancy algorithm which may have attenuated findings with our cumulative vascular
variable.
Obesity was associated with marginally better executive functioning. While this seems
counterintuitive, other HIV cohorts have also reported increased adiposity and BMI are
associated with better global cognitive performance and processing speed and executive
functioning (Gustafson et al., 2013; Lake et al., 2015; Wright et al., 2015). Contradictory
findings in both HIV-seronegative and HIV populations may be in part due to complex, non-
linear relationships with cognition and protective effects of higher BMI later in life (Gustafson,
2008).
53
We also found current smoking were associated with better performance on executive
functioning in PLWH. Previous studies have either reported no relationship between current
smoking and cognition in HIV or detrimental effects of cigarette smoking on motor function
(Fabbiani et al., 2013; Nakamoto et al., 2011; Wright et al., 2010). As mentioned above, CCSE
was inversely associated with cigarette smoking in this study, which is inconsistent with reports
of greater smoking rates in populations with low SES and increased stressor exposure. Further, in
our PLWH group, smokers did not differ from non-smokers on SES, education, and premorbid
estimated intelligence, whereas these differences were found in the HIV- group (data not shown).
Thus, current smokers in the PLWH group may not be representative of larger population, which
may help explain the positive relationship between smoking and executive function.
Alternatively, there is some evidence that cigarette smoking enhances cognition (Valentine &
Sofuoglu, 2018).
Future Directions
One future direction may be to examine sex differences in relationship between stressor
exposure and cognition, and stressor exposure and vascular risk. There is some evidence that
women may be more susceptible to effects of stress on cognition and vascular risk. For example,
in the Framingham heart study, higher cortisol was associated with lower brain volume in
women, but not men (Echouffo-Tcheugui et al., 2018). Further, a review study showed females
may experience greater detrimental stress-related effects on glucose regulation, dyslipidemia,
overall CVD risk, and diet (Taylor et al., 2018). Several studies of stress and the brain in the HIV
literature have performed studies in either all-male or all-female cohorts. Examination of both
sexes within the same design will be helpful in discerning sex differences that are not
confounded by differences between study demographics and methods.
54
Another potential future direction is to examine biological measures of chronic stress
(e.g., cortisol from hair, inflammatory markers) to better understand the biological underpinnings
of cumulative chronic stress and its relationship to the STRAIN.
Strengths
The current study had several strengths. First, this is the first study to our knowledge to
study relationships between lifetime stressor exposure and cognition and brain structure in an
adult sample filling an important gap in the stress literature. There is a general paucity of studies
evaluating cumulative stressor exposure in adults, and lifetime stressor quantification is very
rare. Within the STRAIN, our analyses showed that early life and adulthood stress was
significantly correlated with a moderate effect size demonstrating the importance of looking at
lifetime stress, instead of just early life stress or recent stress. The majority of studies in the
literature have utilized the perceived stress scale, which measures psychological distress within
the past month and does not incorporate any information about exposure to stressful events or
duration. In the few studies that utilized stressor exposure measures, most studies look at stressor
exposure within a short time-frame (e.g., within a year). Thus, the study is very unique in that it
captures adverse life events from birth to present age, and represents an important contribution to
literature on cumulative life stress and health. The mediation analysis is also a strength of this
paper in that it investigates relationships between CCSE, cognition, and brain volume in the
same model allowing us to look at unique contributions of cumulative stressor exposure and
brain volumes to cognitive performance. In addition, our careful analysis of relationships
between stressor exposure and vascular risk is novel in cognitive studies. Lastly, to our
knowledge, only one study has assessed the effects of stressors on cognition over time in PLWH,
55
and this was specific to early life stress (Spies et al., 2017). Thus, the longitudinal design is a
novel to the extant literature.
Limitations
There are also several limitations worth noting. While we controlled for depressive
symptoms, anxiety and PTSD symptoms were not assessed and therefore not included in the
model. It is possible that results may have been attenuated if controlling for these variables. With
regard to longitudinal analyses, the sample size and sample demographics for participants who
completed the follow-up visit represents a limitation.
There are also several limitations worth noting with regard to measurement of vascular
risk. While the Framingham Heart Study algorithm scores have been shown to be an appropriate
measure of risk for heart disease in many studies and populations, and related to both cognition
and cognitive decline, this measure may be less accurate in PLWH. In PLWH, the Framingham
Heart Study 10-year risk for CVD score has been shown to underestimate cardiac events in those
with longer ART use (Law et al., 2006), have poor agreement with subclinical atherosclerosis as
measured by carotid intima media thickness (Parra et al., 2010), and be less strongly associated
with carotid plaques (Monroe et al., 2016). While there are vascular risk burden scores that have
been specifically designed for PLWH, we needed to use a score that could be used in both
controls and PLWH. Furthermore, we lacked certain data to use other algorithms used in the HIV
literature such as family history of cardiovascular disease and biological measurement of risk
factors (e.g., total cholesterol, LDL cholesterol, HDL cholesterol). Furthermore, the modified
FHSCVD measure that uses BMI as a substitute for cholesterol data may be even less
appropriate for PLWH given the high rates of dyslipidemia in PWLH in part due to ART side
effects. It is possible that BMI is an inferior proxy for cholesterol in PLWH. In our sample,
56
PLWH had significantly greater rate of dyslipidemia compared to controls, however there was no
difference in BMI or rates of obesity. Additional exploration looking at relationships between
BMI and dyslipidemia in PLWH and HIV- groups individually showed BMI was associated with
dyslipidemia in the HIV- group but not PLWH. This suggests that cumulative vascular risk may
be underestimated in PLWH in the current sample using the modified algorithm.
Conclusion
This is the first paper to demonstrate lifetime chronic stressor exposure was associated
with worse global cognition and processing speed and executive functioning in PLWH and HIV-
adults. Stressor exposure was inversely associated with brain volumes in the PFC, but not
subcortical structures involved in regulation of HPA axis. Contrary to hypotheses, there were no
significant interactions with HIV status suggesting that cumulative stressor exposure is harmful
to the brain regardless of HIV infection. Mediation analysis showed stressor-related deficits in
processing speed is attributed to reduced PFC volume. Contrary to hypotheses, cumulative
chronic stress was not associated with vascular risk and did not mediate or moderate
relationships between chronic stress and cognition. However, continued exploration of combined
effects of stress and vascular risk on brain function is encouraged given the limitations of our
vascular risk assessment.
This study also has important clinical implications for populations with increased rates of
stressor exposure. Findings warrant development of novel psychosocial interventions to address
the harmful effects of stressor exposure on the brain with the aim of reducing cognitive
dysfunction and potential premature brain aging. Alternatively, existing therapies targeting stress
and depressive/anxiety disorders may be helpful in reducing brain insult. It is possible that
reductions in the PFC associated with chronic stress impede its role in negative feedback to the
57
stress response, which may in turn may decrease emotion regulation and increase sensitivity to
stress.
Overall, findings support the importance of examining cumulative stressor exposure
when studying how stress impacts cognition and brain structure. Future studies may benefit from
larger sample sizes, exploration of gender interactions, and more extensive cognitive follow-up
duration.

58
References
Aartsen, M. J., Van Tilburg, T., Smits, C. H., Comijs, H. C., & Knipscheer, K. C. (2005). Does
widowhood affect memory performance of older persons? Psychological medicine, 35(2),
217-226. https://doi.org/10.1017/s0033291704002831
Aggarwal, N. T., Clark, C. J., Beck, T. L., Mendes de Leon, C. F., DeCarli, C., Evans, D. A., &
Everson Rose, S. A. (2014). Perceived stress is associated with subclinical
cerebrovascular disease in older adults. Am J Geriatr Psychiatry, 22(1), 53-62.
https://doi.org/10.1016/j.jagp.2012.06.001
Aggarwal, N. T., Wilson, R. S., Beck, T. L., Rajan, K. B., Mendes de Leon, C. F., Evans, D. A.,
& Everson-Rose, S. A. (2014). Perceived stress and change in cognitive function among
adults 65 years and older. Psychosom Med, 76(1), 80-85.
https://doi.org/10.1097/psy.0000000000000016
Andel, R., Crowe, M., Kåreholt, I., Wastesson, J., & Parker, M. G. (2011). Indicators of job
strain at midlife and cognitive functioning in advanced old age. The Journals of
Gerontology: Series B, 66(3), 287-291. https://doi.org/10.1093/geronb/gbq105
Andersen, S. L., Tomada, A., Vincow, E. S., Valente, E., Polcari, A., & Teicher, M. H. (2008).
Preliminary evidence for sensitive periods in the effect of childhood sexual abuse on
regional brain development. J Neuropsychiatry Clin Neurosci, 20(3), 292-301.
https://doi.org/10.1176/appi.neuropsych.20.3.29210.1176/jnp.2008.20.3.292
Ansell, E. B., Rando, K., Tuit, K., Guarnaccia, J., & Sinha, R. (2012). Cumulative adversity and
smaller gray matter volume in medial prefrontal, anterior cingulate, and insula regions.
Biol Psychiatry, 72(1), 57-64. https://doi.org/10.1016/j.biopsych.2011.11.022
Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortex structure and
function. Nat Rev Neurosci, 10(6), 410-422. https://doi.org/10.1038/nrn2648
59
Baddeley, A. D., & Hitch, G. (1974). Working memory. Psychology of Learning and Motivation,
8, 47-89. https://doi.org/https://doi.org/10.1016/S0079-7421(08)60452-1
Badran, A., Hollocks, M. J., Brookes, R. L., Morris, R. G., & Markus, H. S. (2019). Framingham
vascular age is associated with worse cognitive performance in the middle-aged and
elderly [research-article]. Aging, Neuropsychology, and Cognition, 26(4).
https://doi.org/https://doi.org/10.1080/13825585.2018.1499866
Barnes, L. L., Lewis, T. T., Begeny, C. T., Yu, L., Bennett, D. A., & Wilson, R. S. (2012).
Perceived Discrimination and Cognition in Older African Americans. Journal of the
International Neuropsychological Society, 18(5), 856-865.
https://doi.org/10.1017%2FS1355617712000628
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-II.
Psychological Corporation.
Belkic, K. L., Landsbergis, P. A., Schnall, P. L., & Baker, D. (2004). Is job strain a major source
of cardiovascular disease risk? Scandinavian journal of work, environment & health,
30(2), 85-128. https://doi.org/10.5271/sjweh.769
Benedict, R. H. B., Schretlen, D., Groninger, L., & Brandt, J. (1998). Hopkins Verbal Learning
Test – Revised: Normative Data and Analysis of Inter-Form and Test-Retest Reliability
[research-article]. The Clinical Neuropsychologist, 12(1).
https://doi.org/https://doi.org/10.1076/clin.12.1.43.1726
Benedict, R. H. B., Schretlen, D., Groninger, L., Dobraski, M., & Shpritz, B. (1996). Revision of
the Brief Visuospatial Memory Test: Studies of normal performance, reliability, and
validity. Psychological Assessment, 8(2), 145. https://doi.org/10.1037/1040-3590.8.2.145
60
Bloss, E. B., Janssen, W. G., McEwen, B. S., & Morrison, J. H. (2010). Interactive effects of
stress and aging on structural plasticity in the prefrontal cortex. J Neurosci, 30(19), 6726-
6731. https://doi.org/10.1523/jneurosci.0759-10.2010
Brandes, D., Ben-Schachar, G., Gilboa, A., Bonne, O., Freedman, S., & Shalev, A. Y. (2002).
PTSD symptoms and cognitive performance in recent trauma survivors. Psychiatry Res,
110(3), 231-238.
Brown, S. M., Henning, S., & Wellman, C. L. (2005). Mild, short-term stress alters dendritic
morphology in rat medial prefrontal cortex. Cereb Cortex, 15(11), 1714-1722.
https://doi.org/10.1093/cercor/bhi048
Cargill, V. A., & Stone, V. E. (2005). HIV/AIDS: a minority health issue. The Medical clinics of
North America, 89(4). https://doi.org/10.1016/j.mcna.2005.03.005
Caswell, L. W., Vitaliano, P. P., Croyle, K. L., Scanlan, J. M., Zhang, J., & Daruwala, A. (2003).
Negative associations of chronic stress and cognitive performance in older adult spouse
caregivers. Exp Aging Res, 29(3), 303-318. https://doi.org/10.1080/03610730303721
Cazassa, M. J., Oliveira, M. D. S., Spahr, C. M., Shields, G. S., & Slavich, G. M. (2020). The
Stress and Adversity Inventory for Adults (Adult STRAIN) in Brazilian Portuguese:
Initial Validation and Links With Executive Function, Sleep, and Mental and Physical
Health. Frontiers in psychology, 10. https://doi.org/10.3389/fpsyg.2019.03083
Cechetto, D. F., Hachinski, V., & Whitehead, S. N. (2008). Vascular risk factors and Alzheimer's
disease. Expert Rev Neurother, 8(5), 743-750. https://doi.org/10.1586/14737175.8.5.743
Cerqueira, J. J., Mailliet, F., Almeida, O. F., Jay, T. M., & Sousa, N. (2007). The prefrontal
cortex as a key target of the maladaptive response to stress. J Neurosci, 27(11), 2781-
2787. https://doi.org/10.1523/jneurosci.4372-06.2007
61
Chow, F. C., Lyass, A., Mahoney, T. F., Massaro, J. M., Triant, V. A., Wu, K., Berzins, B.,
Robertson, K., Ellis, R. J., Tassiopoulos, K., Taiwo, B., & D'Agostino, R. B. (2020).
Baseline 10-Year Cardiovascular Risk Scores Predict Cognitive Function in Older
Persons, and Particularly Women, Living With Human Immunodeficiency Virus
Infection. Clinical infectious diseases : an official publication of the Infectious Diseases
Society of America, 71(12), 3079-3085. https://doi.org/10.1093/cid/ciz1214
Ciccarelli, N., Grima, P., Fabbiani, M., Baldonero, E., Borghetti, A., Milanini, B., Limiti, S.,
Colafigli, M., Tamburrini, E., Cauda, R., De Luca, A., & Di Giambenedetto, S. (2015).
Baseline CD4(+) T-cell count and cardiovascular risk factors predict the evolution of
cognitive performance during 2-year follow-up in HIV-infected patients. Antiviral
Therapy, 20(4), 433-440. https://doi.org/10.3851/imp2925
Clark, U. S., Arce Renteria, M., Hegde, R. R., & Morgello, S. (2018). Early Life Stress-Related
Elevations in Reaction Time Variability Are Associated with Brain Volume Reductions
in HIV+ Adults. Front Behav Neurosci, 12, 6. https://doi.org/10.3389/fnbeh.2018.00006
Clark, U. S., Cohen, R. A., Sweet, L. H., Gongvatana, A., Devlin, K. N., Hana, G. N.,
Westbrook, M. L., Mulligan, R. C., Jerskey, B. A., White, T. L., Navia, B., & Tashima,
K. T. (2012). Effects of HIV and Early Life Stress on Amygdala Morphometry and
Neurocognitive Function. J Int Neuropsychol Soc, 18(4), 657-668.
https://doi.org/10.1017/s1355617712000434
Cohen, R. A., Grieve, S., Hoth, K. F., Paul, R. H., Sweet, L., Tate, D., Gunstad, J., Stroud, L.,
McCaffery, J., Hitsman, B., Niaura, R., Clark, C. R., McFarlane, A., Bryant, R., Gordon,
E., & Williams, L. M. (2006). Early life stress and morphometry of the adult anterior
cingulate cortex and caudate nuclei. Biol Psychiatry, 59(10), 975-982.
https://doi.org/10.1016/j.biopsych.2005.12.016
62
Cohen, R. A., Poppas, A., Forman, D. E., Hoth, K. F., Haley, A. P., Gunstad, J., Jefferson, A. L.,
Tate, D. F., Paul, R. H., Sweet, L. H., Ono, M., Jerskey, B. A., & Gerhard-Herman, M.
(2009). Vascular and cognitive functions associated with cardiovascular disease in the
elderly. J Clin Exp Neuropsychol, 31(1), 96-110. https://doi.org/10.1080/13803390802014594
Cohen, S., Janicki-Deverts, D., & Miller, G. E. (2007). Psychological stress and disease. Jama,
298(14), 1685-1687. https://doi.org/10.1001/jama.298.14.1685
Cohen, S., Murphy, M. L. M., & Prather, A. A. (2019). Ten surprising facts about stressful life
events and disease risk. Annual Review of Psychology, 70, 577-597.
https://doi.org/10.1146/annurev-psych-010418-102857
Cole, J. H., Underwood, J., Caan, M. W., De Francesco, D., van Zoest, R. A., Leech, R., Wit, F.
W., Portegies, P., Geurtsen, G. J., Schmand, B. A., Schim van der Loeff, M. F.,
Franceschi, C., Sabin, C. A., Majoie, C. B., Winston, A., Reiss, P., Sharp, D. J.,
Schouten, J., Kooij, K., Elsenga, B., Janssen, F., Heidenrijk, M., Schrijver, J.,
Zikkenheiner, W., van der Valk, M., Henderiks, A., Kootstra, N., Harskamp-Holwerda,
A., Maurer, I., Ruiz, M. M., Booiman, T., Girigorie, A., Villaudy, J., Frankin, E.,
Pasternak, A., Berkhout, B., van der Kuyl, T., ter Stege, J., Twennaar, M. K., Su, T.,
Siteur-Van Rijnstra, E., Weijer, K., Bisschop, P., Kalsbeek, A., Wezel, M., Visser, I.,
Ruhé, H., Tembo, L., Stott, M., Legg, K., Lovell, A., Erlwein, O., Doyle, N., Norsworthy,
P., Garagnani, P., Pirazzini, C., Capri, M., Dall’Olio, F., Chiricolo, M., Salviol, S.,
Hoeijmakers, J., Pothof, J., Burger, D., de Graaff-Teulen, M., Bürkle, A., Sindlinger, T.,
Moreno-Villanueva, M., Keller, A., Prins, M., Martens, M., Moll, S., Berkel, J., Möller,
L., Visser, G., Zaheri, S., Hillebregt, M., Ruijs, Y., Benschop, D., Gisslén, M., Fuchs, D.,
Zetterberg, H., Libert, C., Dewaele, S., & Guaraldi, G. (2017). Increased brain-predicted
63
aging in treated HIV disease. Neurology, 88(14), 1349-1357.
https://doi.org/10.1212/wnl.0000000000003790
Colodro-Conde, L., Couvy-Duchesne, B., Zhu, G., Coventry, W. L., Byrne, E. M., Gordon, S.,
Wright, M. J., Montgomery, G. W., Madden, P. A. F., Ripke, S., Eaves, L. J., Heath, A.
C., Wray, N. R., Medland, S. E., & Martin, N. G. (2018). A direct test of the diathesis-
stress model for depression. Molecular psychiatry, 23(7), 1590-1596.
https://doi.org/10.1038/mp.2017.130
Comijs, H. C., van den Kommer, T. N., Minnaar, R. W., Penninx, B. W., & Deeg, D. J. (2011).
Accumulated and differential effects of life events on cognitive decline in older persons:
depending on depression, baseline cognition, or ApoE epsilon4 status? The journals of
gerontology. Series B, Psychological sciences and social sciences, 66, 111-120.
https://doi.org/10.1093/geronb/gbr019
Crystal, H. A., Weedon, J., Holman, S., Manly, J., Valcour, V., Cohen, M., Anastos, K., Liu, C.
L., Mack, W. J., Golub, E., Lazar, J., Ho, A., Kreek, M. J., & Kaplan, R. C. (2011).
Associations of cardiovascular variables and HAART with cognition in middle-aged
HIV-infected and uninfected women. Journal of Neurovirology, 17(5), 469-476.
https://doi.org/10.1007/s13365-011-0052-3
D'Agostino, R. B., Sr., Vasan, R. S., Pencina, M. J., Wolf, P. A., Cobain, M., Massaro, J. M., &
Kannel, W. B. (2008). General cardiovascular risk profile for use in primary care: the
Framingham Heart Study. Circulation, 117(6), 743-753.
https://doi.org/10.1161/circulationaha.107.699579
De Socio, G. V., Pasqualini, M., Ricci, E., Maggi, P., Orofino, G., Squillace, N., Menzaghi, B.,
Madeddu, G., Taramasso, L., Francisci, D., Bonfanti, P., Vichi, F., dell'Omo, M., &
Pieroni, L. (2020). Smoking habits in HIV-infected people compared with the general
64
population in Italy: a cross-sectional study. BMC public health, 20(1).
https://doi.org/10.1186/s12889-020-08862-8
Demir, O. M., Candilio, L., Fuster, D., Muga, R., Barbaro, G., Colombo, A., & Azzalini, L.
(2018). Cardiovascular disease burden among human immunodeficiency virus-infected
individuals. Int J Cardiol, 265, 195-203. https://doi.org/10.1016/j.ijcard.2018.03.137
Desikan, R. S., Segonne, F., Fischl, B., Quinn, B. T., Dickerson, B. C., Blacker, D., Buckner, R.
L., Dale, A. M., Maguire, R. P., Hyman, B. T., Albert, M. S., & Killiany, R. J. (2006). An
automated labeling system for subdividing the human cerebral cortex on MRI scans into
gyral based regions of interest. Neuroimage, 31(3), 968-980.
https://doi.org/10.1016/j.neuroimage.2006.01.021
Dikmen, S. S., Heaton, R. K., Grant, I., & Temkin, N. R. (1999). Test-retest reliability and
practice effects of expanded Halstead-Reitan Neuropsychological Test Battery. Journal
of the International Neuropsychological Society, 5(4).
Dregan, A., Stewart, R., & Gulliford, M. C. (2013). Cardiovascular risk factors and cognitive
decline in adults aged 50 and over: a population-based cohort study. Age Ageing, 42(3),
338-345. https://doi.org/10.1093/ageing/afs166
Duron, E., & Hanon, O. (2008). Vascular risk factors, cognitive decline, and dementia. Vasc
Health Risk Manag, 4(2), 363-381.
Echouffo-Tcheugui, J. B., Conner, S. C., Himali, J. J., Maillard, P., DeCarli, C. S., Beiser, A. S.,
Vasan, R. S., & Seshadri, S. (2018). Circulating cortisol and cognitive and structural
brain measures: The Framingham Heart Study. Neurology, 91(21).
https://doi.org/10.1212/WNL.0000000000006549
65
Eckert, M. A., Keren, N. I., Roberts, D. R., Calhoun, V. D., & Harris, K. C. (2010). Age-related
changes in processing speed: unique contributions of cerebellar and prefrontal cortex.
Frontiers in human neuroscience, 4. https://doi.org/10.3389/neuro.09.010.2010
Everson, S. A., Lynch, J. W., Kaplan, G. A., Lakka, T. A., Sivenius, J., & Salonen, J. T. (2001).
Stress-induced blood pressure reactivity and incident stroke in middle-aged men. Stroke,
32(6), 1263-1270.
Fabbiani, M., Ciccarelli, N., Tana, M., Farina, S., Baldonero, E., Di Cristo, V., Colafigli, M.,
Tamburrini, E., Cauda, R., Silveri, M. C., Grima, P., & Di Giambenedetto, S. (2013).
Cardiovascular risk factors and carotid intima-media thickness are associated with lower
cognitive performance in HIV-infected patients. Hiv Medicine, 14(3), 136-144.
https://doi.org/10.1111/j.1468-1293.2012.01044.x
Faeh, D., Braun, J., & Bopp, M. (2012). Body mass index vs cholesterol in cardiovascular
disease risk prediction models. Archives of internal medicine, 172(22), 1766-1768.
https://doi.org/10.1001/2013.jamainternmed.327
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995). The Structured Clinical
Interview for DSM-III-R Personality Disorders (SCID-II): II. Multi-site test-retest
reliability study. Journal of Personality Disorders, 9(2), 92-104.
https://doi.org/10.1521/pedi.1995.9.2.92
Fischl, B., & Dale, A. M. (2000). Measuring the thickness of the human cerebral cortex from
magnetic resonance images. Proc Natl Acad Sci U S A, 97(20), 11050-11055.
https://doi.org/10.1073/pnas.200033797
Fischl, B., Salat, D. H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., van der Kouwe, A.,
Killiany, R., Kennedy, D., Klaveness, S., Montillo, A., Makris, N., Rosen, B., & Dale, A.
66
M. (2002). Whole brain segmentation: automated labeling of neuroanatomical structures
in the human brain. Neuron, 33(3), 341-355. https://doi.org/10.1016/S0896-6273(02)00569-X
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental state". A practical
method for grading the cognitive state of patients for the clinician. J Psychiatr Res, 12(3),
189-198.
Foong, H. F., Hamid, T. A., Ibrahim, R., & Haron, S. A. (2018). Information processing speed as
a mediator between psychosocial stress and global cognition in older adults.
Psychogeriatrics : the official journal of the Japanese Psychogeriatric Society, 18(1), 21-
29. https://doi.org/10.1111/psyg.12279
Frazier, E. L., Sutton, M. Y., Brooks, J. T., Shouse, R. L., & Wesier, J. (2018). Trends in
cigarette smoking among adults with HIV compared with the general adult population,
United States - 2009–2014. Preventive Medicine, 111, 231-234.
https://doi.org/10.1016/j.ypmed.2018.03.007
Gallo, L. C., Bogart, L. M., Vranceanu, A. M., & Walt, L. C. (2004). Job characteristics,
occupational status, and ambulatory cardiovascular activity in women. Annals of
behavioral medicine, 28(1), 62-73. https://doi.org/10.1207/s15324796abm2801_8
Gallo, L. C., Roesch, S. C., Fortmann, A. L., Carnethon, M. R., Penedo, F. J., Perreira, K.,
Birnbaum-Weitzman, O., Wassertheil-Smoller, S., Castañeda, S. F., Talavera, G. A.,
Sotres-Alvarez, D., Daviglus, M. L., Schneiderman, N., & Isasi, C. R. (2014).
Associations of chronic stress burden, perceived stress, and traumatic stress with
cardiovascular disease prevalence and risk factors in the HCHS/SOL Sociocultural
Ancillary Study. Psychosom Med, 76(6), 468-475.
https://doi.org/10.1097/psy.0000000000000069
67
Geronimus, A. T. (1992). The weathering hypothesis and the health of African-American women
and infants: evidence and speculations. Ethnicity & disease, 2(3), 207-221.
Gianaros, P. J., Jennings, J. R., Sheu, L. K., Greer, P. J., Kuller, L. H., & Matthews, K. A.
(2007). Prospective reports of chronic life stress predict decreased grey matter volume in
the hippocampus. Neuroimage, 35(2), 795-803. https://doi.org/10.1016/j.neuroimage.2006.10.045
Golden, C. J., & Freshwater, S. M. (2002). Stroop color and word test: A manual for clinical and
experimental uses. Stoetling Company.
Gomez, D., Power, C., Gill, M. J., & Fujiwara, E. (2017). Determinants of Risk-Taking in HIV-
Associated Neurocognitive Disorders. Neuropsychology, 31(7), 798-810.
https://doi.org/10.1037/neu0000366
Gourley, S. L., Kedves, A. T., Olausson, P., & Taylor, J. R. (2009). A history of corticosterone
exposure regulates fear extinction and cortical NR2B, GluR2/3, and BDNF.
Neuropsychopharmacology, 34(3), 707-716. https://doi.org/10.1038/npp.2008.123
Graham, J. E., Christian, L. M., & Kiecolt-Glaser, J. K. (2006). Stress, Age, and Immune
Function: Toward a Lifespan Approach [ReviewPaper]. Journal of Behavioral Medicine,
29(4), 389-400. https://doi.org/doi:10.1007/s10865-006-9057-4
Grinspoon, S., & Carr, A. (2005). Cardiovascular risk and body-fat abnormalities in HIV-
infected adults. N Engl J Med, 352(1), 48-62. https://doi.org/10.1056/NEJMra041811
Gustafson, D. (2008). A life course of adiposity and dementia. Eur J Pharmacol, 585(1), 163-
175. https://doi.org/10.1016/j.ejphar.2008.01.052
Gustafson, D., Mielke, M. M., Tien, P. C., Valcour, V., Cohen, M., Anastos, K., Liu, C. L.,
Pearce, L., Golub, E. T., Minkoff, H., & Crystal, H. A. (2013). Anthropometric measures
and cognition in middle-aged HIV-infected and uninfected women. The Women's
68
Interagency HIV Study. Journal of Neurovirology, 19(6), 574-585.
https://doi.org/10.1007/s13365-013-0219-1
Hammen, C., Henry, R., & Daley, S. E. (2000). Depression and sensitization to stressors among
young women as a function of childhood adversity. Journal of consulting and clinical
psychology, 68(5), 782-787.
Hanna, D. B., Post, W. S., Deal, J. A., Hodis, H. N., Jacobson, L. P., Mack, W. J., Anastos, K.,
Gange, S. J., Landay, A. L., Lazar, J. M., Palella, F. J., Tien, P. C., Witt, M. D., Xue, X.,
Young, M. A., Kaplan, R. C., & Kingsley, L. A. (2015). HIV Infection Is Associated
With Progression of Subclinical Carotid Atherosclerosis. Clin Infect Dis, 61(4), 640-650.
https://doi.org/10.1093/cid/civ325
Hanson, J. L., Chung, M. K., Avants, B. B., Rudolph, K. D., Shirtcliff, E. A., Gee, J. C.,
Davidson, R. J., & Pollak, S. D. (2012). Structural Variations in Prefrontal Cortex
Mediate the Relationship between Early Childhood Stress and Spatial Working Memory.
Journal of Neuroscience, 32(23), 7919-7925. https://doi.org/10.1523/JNEUROSCI.0307-12.2012
Hanson, J. L., Nacewicz, B. M., Sutterer, M. J., Cayo, A. A., Schaefer, S. M., Rudolph, K. D.,
Shirtcliff, E. A., Pollak, S. D., & Davidson, R. J. (2015). Behavioral problems after early
life stress: contributions of the hippocampus and amygdala. Biological psychiatry, 77(4),
314-323. https://doi.org/10.1016/j.biopsych.2014.04.020
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A
regression-based approach. Guilford Press. https://doi.org/TBD
Heaton, R. K., Clifford, D. B., Franklin, D. R., Jr., Woods, S. P., Ake, C., Vaida, F., Ellis, R. J.,
Letendre, S. L., Marcotte, T. D., Atkinson, J. H., Rivera-Mindt, M., Vigil, O. R., Taylor,
M. J., Collier, A. C., Marra, C. M., Gelman, B. B., McArthur, J. C., Morgello, S.,
Simpson, D. M., McCutchan, J. A., Abramson, I., Gamst, A., Fennema-Notestine, C.,
69
Jernigan, T. L., Wong, J., & Grant, I. (2010). HIV-associated neurocognitive disorders
persist in the era of potent antiretroviral therapy: CHARTER Study. Neurology, 75(23),
2087-2096. https://doi.org/10.1212/WNL.0b013e318200d727
Heaton, R. K., Grant, I., & Mathews, C. G. (1991). Comprehensive norms for an expanded
Halstead-Reitan battery: Demographic corrections, research findings, and clinical
applications; with a supplement for the Wechsler Adult Intelligence Scale-Revised
(WAIS-R). Psychological Assessment Resources.
Heaton, R. K., Marcotte, T. D., Mindt, M. R., Sadek, J., Moore, D. J., Bentley, H., McCutchan, J.
A., Reicks, C., & Grant, I. (2004). The impact of HIV-associated neuropsychological
impairment on everyday functioning. J Int Neuropsychol Soc, 10(3), 317-331.
https://doi.org/10.1017/s1355617704102130
Hedges, D. W., & Woon, F. L. (2011). Early-life stress and cognitive outcome.
Psychopharmacology (Berl), 214(1), 121-130. https://doi.org/10.1007/s00213-010-2090-6
Henderson, K. M., Clark, C. J., Lewis, T. T., Aggarwal, N. T., Beck, T., Guo, H., Lunos, S.,
Brearley, A., Mendes de Leon, C. F., Evans, D. A., & Everson-Rose, S. A. (2013).
Psychosocial distress and stroke risk in older adults. Stroke, 44(2), 367-372.
https://doi.org/10.1161/strokeaha.112.679159
Hergenrather, K. C., Zeglin, R. J., McGuire-Kuletz, M., & Rhodes, S. D. (2015). Employment as
a Social Determinant of Health: A Systematic Review of Longitudinal Studies Exploring
the Relationship Between Employment Status and Physical Health [text]. Rehabilitation,
Research, Policy, and Education, 29(1). https://doi.org/10.1891/2168-6653.29.1.2
Hollingshead, A. B. (1975). Four factor index of social status. Yale University.
70
Holmes, A., & Wellman, C. L. (2009). Stress-induced prefrontal reorganization and executive
dysfunction in rodents. Neurosci Biobehav Rev, 33(6), 773-783.
https://doi.org/10.1016/j.neubiorev.2008.11.005
Jefferson, A. L., Hohman, T. J., Liu, D., Haj-Hassan, S., Gifford, K. A., Benson, E. M., Skinner,
J. S., Lu, Z., Sparling, J., Sumner, E. C., Bell, S., & Ruberg, F. L. (2015). Adverse
vascular risk is related to cognitive decline in older adults. J Alzheimers Dis, 44(4), 1361-
1373. https://doi.org/10.3233/jad-141812
Johnson, A. D., McQuoid, D. R., Steffens, D. C., Payne, M. E., Beyer, J. L., & Taylor, W. D.
(2017). Effects of stressful life events on cerebral white matter hyperintensity
progression. International journal of geriatric psychiatry, 32(12).
https://doi.org/10.1002/gps.4644
Kaijage, T., & Wexler, S. (2010). The Influences of Stress and Social Support on Adherence to
Health Promotion Strategies by People Living with HIV/AIDS in Dar es Salaam,
Tanzania [research-article]. Journal of HIV/AIDS & Social Services, 9(3), 220-239.
https://doi.org/http://dx.doi.org/10.1080/15381501.2010.502807
Kennedy, K. M., & Raz, N. (2009). Aging white matter and cognition: differential effects of
regional variations in diffusion properties on memory, executive functions, and speed.
Neuropsychologia, 47(3), 916-927. https://doi.org/10.1016/j.neuropsychologia.2009.01.001
Kirk, J. B., & Goetz, M. B. (2009). Human immunodeficiency virus in an aging population, a
complication of success. J Am Geriatr Soc, 57(11), 2129-2138. https://doi.org/10.1111/j.1532-
5415.2009.02494.x
Knopman, D., Boland, L. L., Mosley, T., Howard, G., Liao, D., Szklo, M., McGovern, P., &
Folsom, A. R. (2001). Cardiovascular risk factors and cognitive decline in middle-aged
adults. Neurology, 56(1), 42-48.
71
Kuhn, T., Kaufmann, T., Doan, N. T., Westlye, L. T., Jones, J., Nunez, R. A., Bookheimer, S. Y.,
Singer, E. J., Hinkin, C. H., & Thames , A. D. (2018). An augmented aging process in
brain white matter in HIV. Human brain mapping, 39(6). https://doi.org/10.1002/hbm.24019
Kurtzman, K., O'Donovan, A., Koslov, K., Arenander, J., Epel, E. S., & Slavich, G. (2012).
Sweating the big stuff: dispositional pessimism exacerbates the deleterious effects of life
stress on metabolic health. European Journal of Psychotraumatology, Sweden.
Lagraauw, H. M., Kuiper, J., & Bot, I. (2015). Acute and chronic psychological stress as risk
factors for cardiovascular disease: Insights gained from epidemiological, clinical and
experimental studies. Brain Behav Immun, 50, 18-30. https://doi.org/10.1016/j.bbi.2015.08.007
Lake, J. E., Vo, Q. T., Jacobson, L. P., Sacktor, N., Miller, E. N., Post, W. S., Becker, J. T.,
Palella, F. J., Ragin, A., Martin, E., Munro, C. A., & Brown, T. T. (2015). Adiponectin
and interleukin-6, but not adipose tissue, are associated with worse neurocognitive
function in HIV-infected men. Antiviral Therapy, 20(2), 235-244.
https://doi.org/10.3851/imp2952
Law, M. G., Friis-Møller, N., El-Sadr, W. M., Weber, R., Reiss, P., D'Arminio Monforte, A.,
Thiébaut, R., Morfeldt, L., De Wit, S., Pradier, C., Calvo, G., Kirk, O., Sabin, C. A.,
Phillips, A. N., & Lundgren, J. D. (2006). The use of the Framingham equation to predict
myocardial infarctions in HIV-infected patients: comparison with observed events in the
D:A:D Study. HIV medicine, 7(4). https://doi.org/10.1111/j.1468-1293.2006.00362.x
Lazarus, R., & Folkman, S. (1984). Stress, Appraisal, and Coping. Springer.
Leserman, J., Ironson, G., O'Cleirigh, C., Fordiani, J. M., & Balbin, E. (2008). Stressful life
events and adherence in HIV. AIDS Patient Care STDS, 22(5), 403-411.
https://doi.org/10.1089/apc.2007.0175
72
Levy, M. E., Anastos, K., Levine, S. R., Plankey, M., Castel, A. D., Molock, S., Sen, S., Asch, F.
M., Milam, J., Aouizerat, B., Weber, K. M., Golub, E. T., Kaplan, R. C., & Kassaye, S.
(2020). Depression and Psychosocial Stress Are Associated With Subclinical Carotid
Atherosclerosis Among Women Living With HIV. Journal of the American Heart
Association, 9(13). https://doi.org/10.1161/JAHA.120.016425
Liang, Y. L., Shiel, L. M., Teede, H., Kotsopoulos, D., McNeil, J., Cameron, J. D., & McGrath,
B. P. (2001). Effects of Blood Pressure, Smoking, and Their Interaction on Carotid
Artery Structure and Function. Hypertension, 37(1). https://doi.org/10.1161/01.hyp.37.1.6
Liu, R. J., & Aghajanian, G. K. (2008). Stress blunts serotonin- and hypocretin-evoked EPSCs in
prefrontal cortex: role of corticosterone-mediated apical dendritic atrophy. Proc Natl
Acad Sci U S A, 105(1), 359-364. https://doi.org/10.1073/pnas.0706679105
Lupien, S., Lecours, A. R., Lussier, I., Schwartz, G., Nair, N. P., & Meaney, M. J. (1994). Basal
cortisol levels and cognitive deficits in human aging. J Neurosci, 14(5 Pt 1), 2893-2903.
Lupien, S. J., de Leon, M., de Santi, S., Convit, A., Tarshish, C., Nair, N. P., Thakur, M.,
McEwen, B. S., Hauger, R. L., & Meaney, M. J. (1998). Cortisol levels during human
aging predict hippocampal atrophy and memory deficits. Nat Neurosci, 1(1), 69-73.
https://doi.org/10.1038/271
Lupien, S. J., Juster, R. P., Raymond, C., & Marin, M. F. (2018). The effects of chronic stress on
the human brain: From neurotoxicity, to vulnerability, to opportunity. Frontiers in
neuroendocrinology, 49, 91-105. https://doi.org/10.1016/j.yfrne.2018.02.001
Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout the
lifespan on the brain, behaviour and cognition. Nature reviews. Neuroscience, 10(6), 434-
445. https://doi.org/10.1038/nrn2639
73
Machtinger, E. L., Wilson, T. C., Haberer, J. E., & Weiss, D. S. (2012). Psychological trauma
and PTSD in HIV-positive women: a meta-analysis. AIDS and behavior, 16(8).
https://doi.org/10.1007/s10461-011-0127-4
Maki, P. M., Rubin, L. H., Valcour, V., Martin, E., Crystal, H., Young, M., Weber, K. M.,
Manly, J., Richardson, J., Alden, C., & Anastos, K. (2015). Cognitive function in women
with HIV: findings from the Women's Interagency HIV Study. Neurology, 84(3), 231-
240. https://doi.org/10.1212/WNL.0000000000001151
Marin, M. F., Lord, C., Andrews, J., Juster, R. P., Sindi, S., Arsenault-Lapierre, G., Fiocco, A. J.,
& Lupien, S. J. (2011). Chronic stress, cognitive functioning and mental health.
Neurobiol Learn Mem, 96(4), 583-595. https://doi.org/10.1016/j.nlm.2011.02.016
Mauss, D., Volmer-Thole, M., Herr, R., Bosch, J. A., & Fischer, J. E. (2020). Stress at work is
associated with intima media thickness in older male employees, independent of other
sources of stress perception. Stress. https://doi.org/10.1080/10253890.2020.1812059
McEwen, B. S. (2007). Physiology and neurobiology of stress and adaptation: central role of the
brain. Physiological reviews, 87(3). https://doi.org/10.1152/physrev.00041.2006
McEwen, B. S., Nasca, C., & Gray, J. D. (2016). Stress Effects on Neuronal Structure:
Hippocampus, Amygdala, and Prefrontal Cortex. Neuropsychopharmacology : official
publication of the American College of Neuropsychopharmacology, 41(1), 3-23.
https://doi.org/10.1038/npp.2015.171
McEwen, B. S., & Stellar, E. (1993). Stress and the individual. Mechanisms leading to disease.
Archives of internal medicine, 153(18), 2093-2101.
McGonagle, K. A., & Kessler, R. C. (1990). Chronic stress, acute stress, and depressive
symptoms. American journal of community psychology, 18(5), 681-706.
https://doi.org/10.1007/BF00931237
74
McIntosh, E. C., Tureson, K., Rotblatt, L. J., Singer, E. J., & Thames, A. D. (2021). HIV,
Vascular Risk Factors, and Cognition in the Combination Antiretroviral Therapy Era: A
Systematic Review and Meta-Analysis. Journal of the International Neuropsychological
Society, 27(4), 365-381. https://doi.org/10.1017/S1355617720001022
McLoughlin, E., Fletcher, D., Slavich, G. M., Arnold, R., & Moore, L. J. (2021). Cumulative
lifetime stress exposure, depression, anxiety, and well-being in elite athletes: A mixed-
method study. Psychology of Sport and Exercise, 52.
https://doi.org/10.1016/j.psychsport.2020.101823
Mdodo, R., Frazier, E. L., Dube, S. R., Mattson, C. L., Sutton, M. Y., Brooks, J. T., &
Skarbinski, J. (2015). Cigarette smoking prevalence among adults with HIV compared
with the general adult population in the United States: cross-sectional surveys. Ann Intern
Med, 162(5), 335-344. https://doi.org/10.7326/m14-0954
Mehta, M. A., Golembo, N. I., Nosarti, C., Colvert, E., Mota, A., Williams, S. C., Rutter, M., &
Sonuga-Barke, E. J. (2009). Amygdala, hippocampal and corpus callosum size following
severe early institutional deprivation: the English and Romanian Adoptees study pilot. J
Child Psychol Psychiatry, 50(8), 943-951. https://doi.org/10.1111/j.1469-7610.2009.02084.x
Miller, L. S., & Rohling, M. L. (2001). A statistical interpretive method for neuropsychological
test data. Neuropsychology review, 11(3). https://doi.org/10.1023/a:1016602708066
Mitchell, C. G., & Linsk, N. L. (2004). A multidimensional conceptual framework for
understanding HIV/AIDS as a chronic long-term illness. Soc Work, 49(3), 469-477.
Monroe, A. K., Haberlen, S. A., Post, W. S., Palella, F. J., Kinsgley, L. A., Witt, M. D., Budoff,
M., Jacobson, L. P., & Brown, T. T. (2016). Cardiovascular disease risk scores'
relationship to subclinical cardiovascular disease among HIV-infected and HIV-
uninfected men. AIDS, 30(13), 2075-2084. https://doi.org/10.1097/QAD.0000000000001163
75
Moreno, G. L., Bruss, J., & Denburg, N. L. (2017). Increased perceived stress is related to
decreased prefrontal cortex volumes among older adults. Journal of clinical and
experimental neuropsychology, 39(4), 313-325. https://doi.org/10.1080/13803395.2016.1225006
Mukerji, S. S., Locascio, J. J., Misra, V., Lorenz, D. R., Holman, A., Dutta, A., Penugonda, S.,
Wolinsky, S. M., & Gabuzda, D. (2016). Lipid Profiles and APOE4 Allele Impact
Midlife Cognitive Decline in HIV-Infected Men on Antiretroviral Therapy. Clin Infect
Dis, 63(8), 1130-1139. https://doi.org/10.1093/cid/ciw495
Munoz, E., Sliwinski, M. J., Scott, S. B., & Hofer, S. (2015). Global perceived stress predicts
cognitive change among older adults. Psychology and aging, 30(3), 487-499.
https://doi.org/10.1037/pag0000036
Nakamoto, B. K., Valcour, V., Kallianpur, K., Liang, C. Y., McMurtray, A., Chow, D.,
Kappenburg, E., & Shikuma, C. M. (2011). Impact of Cerebrovascular Disease on
Cognitive Function in HIV-infected Patients. J Acquir Immune Defic Syndr, 57(3), e66-
68. https://doi.org/10.1097/QAI.0b013e31821ff8bd
O'Neal, W. T., Qureshi, W., Judd, S. E., Glasser, S. P., Ghazi, L., Pulley, L., Howard, V. J.,
Howard, G., & Soliman, E. Z. (2015). Perceived Stress and Atrial Fibrillation: The
REasons for Geographic and Racial Differences in Stroke Study. Ann Behav Med, 49(6),
802-808. https://doi.org/10.1007/s12160-015-9715-2
Ohman, L., Nordin, S., Bergdahl, J., Slunga Birgander, L., & Stigsdotter Neely, A. (2007).
Cognitive function in outpatients with perceived chronic stress. Scand J Work Environ
Health, 33(3), 223-232.
Pakenham, K. (2010). Development of the HIV/AIDS Stress Scale [research-article].
https://doi.org/10.1080/08870440290013680. https://doi.org/Psychology and Health, Vol. 17, No. 2,
2002, pp. 203-219
76
Parra, S., Coll, B., Aragonés, G., Marsillach, J., Beltrán, R., Rull, A., Joven, J., Alonso-
Villaverde, C., & Camps, J. (2010). Nonconcordance between subclinical atherosclerosis
and the calculated Framingham risk score in HIV-infected patients: relationships with
serum markers of oxidation and inflammation. HIV medicine, 11(4), 225-231.
https://doi.org/10.1111/j.1468-1293.2009.00766.x
Pascual, R., & Zamora-León, S. P. (2007). Effects of neonatal maternal deprivation and
postweaning environmental complexity on dendritic morphology of prefrontal pyramidal
neurons in the rat. Acta neurobiologiae experimentalis, 67(4).
Paul, R., Henry, L., Grieve, S. M., Guilmette, T. J., Niaura, R., Bryant, R., Bruce, S., Williams,
L. M., Richard, C. C., Cohen, R. A., & Gordon, E. (2008). The relationship between early
life stress and microstructural integrity of the corpus callosum in a non-clinical
population. Neuropsychiatr Dis Treat, 4(1), 193-201.
Peavy, G. M., Salmon, D. P., Jacobson, M. W., Hervey, A., Gamst, A. C., Wolfson, T.,
Patterson, T. L., Goldman, S., Mills, P. J., Khandrika, S., & Galasko, D. (2009). Effects
of chronic stress on memory decline in cognitively normal and mildly impaired older
adults. The American journal of psychiatry, 166(12), 1384-1391.
https://doi.org/10.1176/appi.ajp.2009.09040461
Pence, B. W., Reif, S., Whetten, K., Leserman, J., Stangl, D., Swartz, M., Thielman, N., &
Mugavero, M. J. (2007). Minorities, the poor, and survivors of abuse: HIV-infected
patients in the US deep South. South Med J, 100(11), 1114-1122.
https://doi.org/10.1097/01.smj.0000286756.54607.9f
Piccolo, L. R., & Noble, K. G. (2018). Perceived stress is associated with smaller hippocampal
volume in adolescence. Psychophysiology, 55(5). https://doi.org/10.1111/psyp.13025
77
Pukay-Martin, N. D., Cristiani, S. A., Saveanu, R., & Bornstein, R. A. (2003). The relationship
between stressful life events and cognitive function in HIV-infected men. J
Neuropsychiatry Clin Neurosci, 15(4), 436-441. https://doi.org/10.1176/jnp.15.4.436
Pyykkönen, A. J., Räikkönen, K., Tuomi, T., Eriksson, J. G., Groop, L., & Isomaa, B. (2010).
Stressful life events and the metabolic syndrome: the prevalence, prediction and
prevention of diabetes (PPP)-Botnia Study. Diabetes care, 33(2), 378-384.
https://doi.org/10.2337/dc09-1027
Radin, R. M., Mason, A. E., Laudenslager, M. L., & Epel, E. S. (2019). Maternal caregivers have
confluence of altered cortisol, high reward-driven eating, and worse metabolic health.
PloS one, 14(5). https://doi.org/10.1371/journal.pone.0216541
Radley, J. J., Rocher, A. B., Miller, M., Janssen, W. G., Liston, C., Hof, P. R., McEwen, B. S., &
Morrison, J. H. (2006). Repeated stress induces dendritic spine loss in the rat medial
prefrontal cortex. Cereb Cortex, 16(3), 313-320. https://doi.org/10.1093/cercor/bhi104
Redmond, N., Richman, J., Gamboa, C. M., Albert, M. A., Sims, M., Durant, R. W., Glasser, S.
P., & Safford, M. M. (2013). Perceived stress is associated with incident coronary heart
disease and all-cause mortality in low- but not high-income participants in the Reasons
for Geographic And Racial Differences in Stroke study. J Am Heart Assoc, 2(6),
e000447. https://doi.org/10.1161/jaha.113.000447
Reschke-Hernández, A. E., Okerstrom, K. L., Bowles Edwards, A., & Tranel, D. (2017). Sex and
stress: Men and women show different cortisol responses to psychological stress induced
by the Trier social stress test and the Iowa singing social stress test. Journal of
neuroscience research, 95(1-2), 106-114. https://doi.org/10.1002/jnr.23851
78
Roberts Jr, R. J., & Pennington, B. F. (1996). An interactive framework for examining prefrontal
cognitive processes. Developmental Neuropsychology, 12(1), 105-126.
https://doi.org/10.1080/87565649609540642
Rock, P. L., Roiser, J. P., Riedel, W. J., & Blackwell, A. D. (2014). Cognitive impairment in
depression: a systematic review and meta-analysis. Psychological medicine, 44(10),
2029-2040. https://doi.org/10.1017/S0033291713002535
Rod, N. H., Gronbaek, M., Schnohr, P., Prescott, E., & Kristensen, T. S. (2009). Perceived stress
as a risk factor for changes in health behaviour and cardiac risk profile: a longitudinal
study. J Intern Med, 266(5), 467-475. https://doi.org/10.1111/j.1365-2796.2009.02124.x
Rosnick, C. B., Small, B. J., McEvoy, C. L., Borenstein, A. R., & Mortimer, J. A. (2007).
Negative life events and cognitive performance in a population of older adults. Journal of
aging and health, 19(4). https://doi.org/10.1177/0898264307300975
Rozanski, A., Blumenthal, J. A., & Kaplan, J. (1999). Impact of psychological factors on the
pathogenesis of cardiovascular disease and implications for therapy. Circulation, 99(16),
2192-2217.
Rubin, L. H., Cook, J. A., Weber, K. M., Cohen, M. H., Martin, E., Valcour, V., Milam, J.,
Anastos, K., Young, M. A., Alden, C., Gustafson, D. R., & Maki, P. M. (2015). The
association of perceived stress and verbal memory is greater in HIV-infected versus HIV-
uninfected women. J Neurovirol, 21(4), 422-432. https://doi.org/10.1007/s13365-015-0331-5
Rubin, L. H., Meyer, V. J., R, J. C., Sundermann, E. E., Wu, M., Weber, K. M., Cohen, M. H.,
Little, D. M., & Maki, P. M. (2016). Prefrontal cortical volume loss is associated with
stress-related deficits in verbal learning and memory in HIV-infected women. Neurobiol
Dis, 92(Pt B), 166-174. https://doi.org/10.1016/j.nbd.2015.09.010
79
Rubin, L. H., Wu, M., Sundermann, E. E., Meyer, V. J., Smith, R., Weber, K. M., Cohen, M. H.,
Little, D. M., & Maki, P. M. (2016). Elevated stress is associated with prefrontal cortex
dysfunction during a verbal memory task in women with HIV. Journal of neurovirology,
22(6), 840-851. https://doi.org/10.1007/s13365-016-0446-3
Ruff, R. M., Light, R. H., Parker, S. B., & Levin, H. S. (1996). Benton Controlled Oral Word
Association Test: reliability and updated norms. Archives of clinical neuropsychology,
11(4), 329-338.
Ruff, R. M., & Parker, S. B. (1993). Gender- and age-specific changes in motor speed and eye-
hand coordination in adults: normative values for the Finger Tapping and Grooved
Pegboard Tests. Perceptual and motor skills, 76(3). https://doi.org/10.2466/pms.1993.76.3c.1219
Räikkönen, K., Matthews, K. A., & Kuller, L. H. (2007). Depressive symptoms and stressful life
events predict metabolic syndrome among middle-aged women: a comparison of World
Health Organization, Adult Treatment Panel III, and International Diabetes Foundation
definitions. Diabetes care, 30(4), 872-877. https://doi.org/10.2337/dc06-1857
Sabia, S., Kivimaki, M., Shipley, M. J., Marmot, M. G., & Singh-Manoux, A. (2009). Body mass
index over the adult life course and cognition in late midlife: the Whitehall II Cohort
Study. Am J Clin Nutr, 89(2), 601-607. https://doi.org/10.3945/ajcn.2008.26482
Salthouse, T. A. (2009). When does age-related cognitive decline begin? Neurobiology of aging,
30(4). https://doi.org/10.1016/j.neurobiolaging.2008.09.023
Sapolsky, R. M., Krey, L. C., & McEwen, B. S. (1986). The neuroendocrinology of stress and
aging: the glucocorticoid cascade hypothesis. Endocr Rev, 7(3), 284-301.
https://doi.org/10.1210/edrv-7-3-284
80
Schoofs, D., Pabst, S., Brand, M., & Wolf, O. T. (2013). Working memory is differentially
affected by stress in men and women. Behavioural brain research, 241, 144-153.
https://doi.org/10.1016/j.bbr.2012.12.004
Seaberg, E. C., Benning, L., Sharrett, A. R., Lazar, J. M., Hodis, H. N., Mack, W. J., Siedner, M.
J., Phair, J. P., Kingsley, L. A., & Kaplan, R. C. (2010). Association between human
immunodeficiency virus infection and stiffness of the common carotid artery. Stroke,
41(10), 2163-2170. https://doi.org/10.1161/strokeaha.110.583856
Seckfort, D. L., Paul, R., Grieve, S. M., Vanderberg, B., Bryant, R. A., Williams, L. M., Clark,
C. R., Cohen, R. A., Bruce, S., & Gordon, E. (2008). Early Life Stress on Brain Structure
and Function Across the Lifespan: A Preliminary Study | SpringerLink. Brain Imaging
and Behavior, 2, 49-58. https://doi.org/10.1007/s11682-007-9015-y
Shaked, D., Katzel, L. I., Seliger, S. L., Gullapalli, R. P., Davatzikos, C., Erus, G., Evans, M. K.,
Zonderman, A. B., & Waldstein, S. R. (2018). Dorsolateral prefrontal cortex volume as a
mediator between socioeconomic status and executive function. Neuropsychology, 32(8),
985-995. https://doi.org/10.1037/neu0000484
Shields, G. S., & Slavich, G. M. (2017). Lifetime Stress Exposure and Health: A Review of
Contemporary Assessment Methods and Biological Mechanisms. Social and personality
psychology compass, 11(8). https://doi.org/10.1111/spc3.12335
Simon, M., Czéh, B., & Fuchs, E. (2005). Age-dependent susceptibility of adult hippocampal
cell proliferation to chronic psychosocial stress. Brain research, 1049(2).
https://doi.org/10.1016/j.brainres.2005.05.006
Singer, E. J., Valdes-Sueiras, M., Commins, D. L., Yong, W., & Carlson, M. (2013). HIV stroke
risk: evidence and implications. Ther Adv Chronic Dis, 4(2), 61-70.
https://doi.org/10.1177/2040622312471840
81
Slavich, G. M., & Irwin, M. R. (2014). From stress to inflammation and major depressive
disorder: a social signal transduction theory of depression. Psychological bulletin, 140(3),
774-815. https://doi.org/10.1037/a0035302
Slavich, G. M., & Shields, G. S. (2018). Assessing Lifetime Stress Exposure Using the Stress
and Adversity Inventory for Adults (Adult STRAIN): An Overview and Initial
Validation. Psychosom Med, 80(1), 17-27. https://doi.org/10.1097/psy.0000000000000534
So-Armah, K., & Freiberg, M. (2014). CVD risk in an aging HIV population – not just a
question of biology. Curr Opin HIV AIDS, 9(4), 346-354.
https://doi.org/10.1097/coh.0000000000000065
Spies, G., Ahmed-Leitao, F., Fennema-Notestine, C., Cherner, M., & Seedat, S. (2016). Effects
of HIV and childhood trauma on brain morphometry and neurocognitive function. J
Neurovirol, 22(2), 149-158. https://doi.org/10.1007/s13365-015-0379-2
Spies, G., Fennema-Notestine, C., Cherner, M., & Seedat, S. (2017). Changes in cognitive
function in women with HIV infection and early life stress. AIDS Care, 29(1), 14-23.
https://doi.org/10.1080/09540121.2016.1204417
Spies, G., Mall, S., Wieler, H., Masilela, L., Castelon Konkiewitz, E., & Seedat, S. (2020). The
relationship between potentially traumatic or stressful events, HIV infection and
neurocognitive impairment (NCI): a systematic review of observational epidemiological
studies. European journal of psychotraumatology, 11(1).
https://doi.org/10.1080/20008198.2020.1781432
Stein, M. B., Koverola, C., Hanna, C., Torchia, M. G., & McClarty, B. (1997). Hippocampal
volume in women victimized by childhood sexual abuse. Psychol Med, 27(4), 951-959.
82
Sublette, M. E., Galfalvy, H. C., Oquendo, M. A., Bart, C. P., Schneck, N., Arango, V., & Mann,
J. J. (2016). Relationship of recent stress to amygdala volume in depressed and healthy
adults. Journal of affective disorders, 203, 136-142. https://doi.org/10.1016/j.jad.2016.05.036
Sumner, J. A., Hagan, K., Grodstein, F., Roberts, A. L., Harel, B., & Koenen, K. C. (2017).
Posttraumatic stress disorder symptoms and cognitive function in a large cohort of
middle-aged women. Depress Anxiety, 34(4), 356-366. https://doi.org/10.1002/da.22600
Sussams, R., Schlotz, W., Clough, Z., Amin, J., Simpson, S., Abbott, A., Beardmore, R.,
Sharples, R., Raybould, R., Brookes, K., Morgan, K., Culliford, D., & Holmes, C. (2020).
Psychological stress, cognitive decline and the development of dementia in amnestic mild
cognitive impairment. Scientific reports, 10(1). https://doi.org/10.1038/s41598-020-60607-0
Taylor, J. L., Makarem, N., Shimbo, D., & Aggarwal, B. (2018). Gender Differences in
Associations Between Stress and Cardiovascular Risk Factors and Outcomes [research-
article]. Gender and the Genome, 2(4), 111-122. https://doi.org/10.1177_2470289718820845
Thames, A. D., Arentoft, A., Rivera-Mindt, M., & Hinkin, C. H. (2013). Functional disability in
medication management and driving among individuals with HIV: a 1-year follow-up
study. J Clin Exp Neuropsychol, 35(1), 49-58. https://doi.org/10.1080/13803395.2012.747596
Tottenham, N., Hare, T. A., Quinn, B. T., McCarry, T. W., Nurse, M., Gilhooly, T., Millner, A.,
Galvan, A., Davidson, M. C., Eigsti, I. M., Thomas, K. M., Freed, P. J., Booma, E. S.,
Gunnar, M. R., Altemus, M., Aronson, J., & Casey, B. J. (2010). Prolonged institutional
rearing is associated with atypically large amygdala volume and difficulties in emotion
regulation. Dev Sci, 13(1), 46-61. https://doi.org/10.1111/j.1467-7687.2009.00852.x
Tottenham, N., & Sheridan, M. A. (2009). A review of adversity, the amygdala and the
hippocampus: a consideration of developmental timing. Front Hum Neurosci, 3, 68.
https://doi.org/10.3389/neuro.09.068.2009
83
Triant, V. A. (2013). Cardiovascular disease and HIV infection. Curr HIV/AIDS Rep, 10(3), 199-
206. https://doi.org/10.1007/s11904-013-0168-6
Truelsen, T., Nielsen, N., Boysen, G., & Gronbaek, M. (2003). Self-reported stress and risk of
stroke: the Copenhagen City Heart Study. Stroke, 34(4), 856-862.
https://doi.org/10.1161/01.str.0000062345.80774.40
Tschanz, J. T., Pfister, R., Wanzek, J., Corcoran, C., Smith, K., Tschanz, B. T., Steffens, D. C.,
Østbye, T., Welsh-Bohmer, K. A., & Norton, M. C. (2013). Stressful life events and
cognitive decline in late life: moderation by education and age. The Cache County Study.
International journal of geriatric psychiatry, 28(8), 821-830. https://doi.org/10.1002/gps.3888
Tucker, A. M., & Stern, Y. (2011). Cognitive reserve in aging. Current Alzheimer research, 8(4).
https://doi.org/10.2174/156720511795745320
Tucker-Drob, E. M., Brandmaier, A. M., & Lindenberger, U. (2019). Coupled cognitive changes
in adulthood: A meta-analysis. Psychological Bulletin, 145(3).
https://doi.org/10.1037/bul0000179
Turner, A. D., James, B. D., Capuano, A. W., Aggarwal, N. T., & Barnes, L. L. (2017).
Perceived Stress and Cognitive Decline in Different Cognitive Domains in a Cohort of
Older African Americans. Am J Geriatr Psychiatry, 25(1), 25-34.
https://doi.org/10.1016/j.jagp.2016.10.003
Uddin, L. Q., Nomi, J. S., Hébert-Seropian, B., Ghaziri, J., & Boucher, O. (2017). Structure and
Function of the Human Insula. Journal of clinical neurophysiology, 34(4), 300-306.
https://doi.org/10.1097/WNP.0000000000000377
Valdez, A. N., Rubin, L. H., & Neigh, G. N. (2016). Untangling the Gordian knot of HIV, stress,
and cognitive impairment. Neurobiol Stress, 4, 44-54.
https://doi.org/10.1016/j.ynstr.2016.02.005
84
Valentine, G., & Sofuoglu, M. (2018). Cognitive Effects of Nicotine: Recent Progress. Current
neuropharmacology, 16(4), 403-414. https://doi.org/10.2174/1570159X15666171103152136
Viticchi, G., Falsetti, L., Buratti, L., Sajeva, G., Luzzi, S., Bartolini, M., Provinciali, L., &
Silvestrini, M. (2017). Framingham Risk Score and the Risk of Progression from Mild
Cognitive Impairment to Dementia. Journal of Alzheimer's disease, 59(1).
https://doi.org/10.3233/JAD-170160
Vyas, A., Mitra, R., Shankaranarayana Rao, B. S., & Chattarji, S. (2002). Chronic stress induces
contrasting patterns of dendritic remodeling in hippocampal and amygdaloid neurons. J
Neurosci, 22(15), 6810-6818. https://doi.org/20026655
Walker, R. J., Garacci, E., Campbell, J. A., & Egede, L. E. (2020). The influence of daily stress
on glycemic control and mortality in adults with diabetes. Journal of behavioral
medicine, 43(5). https://doi.org/10.1007/s10865-019-00109-1
Waters, A. B., Mace, R. A., Sawyer, K. S., & Gansler, D. A. (2019). Identifying errors in
Freesurfer automated skull stripping and the incremental utility of manual intervention.
Brain imaging and behavior, 13(5), 1281-1291. https://doi.org/10.1007/s11682-018-9951-8
Wechsler, D. (2008). Wechsler Adult Intelligence Scale--Fourth Edition (WAIS-IV). Pearson.
https://doi.org/10.1037/t15169-000
Wellman, C. L. (2001). Dendritic reorganization in pyramidal neurons in medial prefrontal
cortex after chronic corticosterone administration. J Neurobiol, 49(3), 245-253.
Whitworth, J. A., Williamson, P. M., Mangos, G., & Kelly, J. J. (2005). Cardiovascular
consequences of cortisol excess. Vascular health and risk management, 1(4).
https://doi.org/10.2147/vhrm.2005.1.4.291
Wiernik, E., Nabi, H., Pannier, B., Czernichow, S., Hanon, O., Simon, T., Simon, J. M., Thomas,
F., Ducolombier, C., Danchin, N., Limosin, F., Consoli, S. M., & Lemogne, C. (2014).
85
Perceived stress, sex and occupational status interact to increase the risk of future high
blood pressure: the IPC cohort study. J Hypertens, 32(10), 1979-1986; discussion 1986.
https://doi.org/10.1097/hjh.0000000000000288
Wright, E. J., Grund, B., Cysique, L. A., Robertson, K. R., Brew, B. J., Collins, G., Shlay, J. C.,
Winston, A., Read, T. R. H., Price, R. W., & Int Network Strategic, I. (2015). Factors
associated with neurocognitive test performance at baseline: a substudy of the INSIGHT
Strategic Timing of AntiRetroviral Treatment (START) trial. Hiv Medicine, 16, 97-108.
https://doi.org/10.1111/hiv.12238
Wright, E. J., Grund, B., Robertson, K., Brew, B. J., Roediger, M., Bain, M. P., Drummond, F.,
Vjecha, M. J., Hoy, J., Miller, C., de Oliveira, A. C. P., Pumpradit, W., Shlay, J. C., El-
Sadr, W., Price, R. W., & Grp, I. S. S. (2010). Cardiovascular risk factors associated with
lower baseline cognitive performance in HIV-positive persons. Neurology, 75(10), 864-
873. https://doi.org/10.1212/WNL.0b013e3181f11bd8
Wu, J., Tong, H., Liu, Z., Tao, J., Chen, L., Chan, C. C. H., & Lee, T. M. C. (2021).
Neurobiological effects of perceived stress are different between adolescents and middle-
aged adults. Brain imaging and behavior, 15(2), 846-854. https://doi.org/10.1007/s11682-020-
00294-7
Xavier, F. M., Ferraz, M. P., Trentini, C. M., Freitas, N. K., & Moriguchi, E. H. (2002).
Bereavement-related cognitive impairment in an oldest-old community-dwelling
Brazilian sample. Journal of clinical and experimental neuropsychology, 24(3), 294-301.
https://doi.org/10.1076/jcen.24.3.294.983
Zhu, M. Y., Wang, W. P., Huang, J., & Regunathan, S. (2007). Chronic treatment with
glucocorticoids alters rat hippocampal and prefrontal cortical morphology in parallel with
86
endogenous agmatine and arginine decarboxylase levels. J Neurochem, 103(5), 1811-
1820. https://doi.org/10.1111/j.1471-4159.2007.04867.x
Zimmerman, M. E., Ezzati, A., Katz, M. J., Lipton, M. L., Brickman, A. M., Sliwinski, M. J., &
Lipton, R. B. (2016). Perceived Stress Is Differentially Related to Hippocampal Subfield
Volumes among Older Adults. PLoS One, 11(5), e0154530.
https://doi.org/10.1371/journal.pone.0154530

87
Tables
Exposure
Indices
Exposure
Timing
Stressor
Types
Primary Life
Domains
Core Social-
Psychological
Characteristics
Stressor
count
Early life stress Acute life
events
Housing Interpersonal loss
Stressor
severity
Adulthood life
stress
Chronic
difficulties
Education Physical danger
Work Humiliation
Continuous age
across the life
course
Treatment/Health Entrapment
Marital/Partner Role change/Disruption
Reproduction
Financial
Legal/Crime
Other
Relationships

Death
Life-threatening
situations

Possessions
Table 1. Dimensions of life stress assessed by the Stress and Adversity Inventory for Adults
(Adult STRAIN). Taken from Slavich and Shields, 2018.

88
Whole
Sample
(N = 161)
HIV-
(n = 95)
PLWH
(n = 66)
Statistic P-
Value
Age 56.96
(7.37)
56.77
(7.52)
57.26 (7.15) 0.172 0.679
Gender (% Male) 68.9 64.6 74.2 1.692 0.193
Race (% Black) 63.4 63.5 62.1 0.034 0.854
Education, years 13.92
(2.30)
14.16
(2.34)
13.64 (2.26) 1.981 0.161
WRAT (standard score) 98.71
(16.45)
99.61
(16.36)
97.41
(16.62)
2918.00 0.455
Hepatitis C (%) 4.3 1.1 9.2 6.170 0.013
Socioeconomic Status
Hollingshead Total 40.54
(12.17)
40.32
(12.35)
41.10
(12.03)
0.157 0.692
Household Income* 36.14
(37.44)
44.58
(41.28)
23.28
(26.38)
1611.00 <0.001
Income Per Family* Member 26.92
(23.49)
31.49
(23.56)
20.28
(21.93)
1831.00 <0.001
Psychiatric/Stressors
BDI-2 5.86 (7.13) 4.69
(6.36)
7.59 (7.82) 2412.50 0.012
Current Depression (%) 2.5 2.1 3.0 0.138 0.711
Past Depression (%) 16.8 8.4 28.8 11.574 0.001
Past Mania or Psychosis (%) 4.3 1.1 9.1 4.619 0.032
CCSE 32.63
(22.93)
29.03
(20.49)
37.68
(25.26)
2522.50 0.035
Health-related Stress 8.94 (8.10) 6.52
(6.65)
12.42 (8.75) 1818.001 <0.001
Substance Use
Past Substance Dependance
(%)
32.3 18.9 52.3 19.579 <0.001
Current Marijuana/Alcohol
Abuse (%)
7.5 6.3 9.2 0.473 0.492
HIV Characteristics
Nadir CD4 - - 279.78
(243.75)
- -
Current CD4 732.28
(319.42)

Highest VL (log) - - 2.29 (0.50) - -
cART Use (%) - - 98.5 - -
Years Since Diagnosis 20.77 (8.91)
Table 2. Demographic and clinical characteristics of baseline study sample.
89
Abbreviations: WRAT (Word Reading Association Test); SES (socioeconomic status); BDI =
Beck Depression Inventory; CCSE = cumulative chronic stressor exposure; VL = viral load;
cART = combined antiretroviral therapy
*Values divided by 1000

90
Whole
Sample
(N = 161)
HIV-
(n = 95)
PLWH
(n = 66)
Statistic P-
Value
Vascular Risk Factors
10-year CVD risk 18.47 (9.10) 17.37 (9.23) 19.80 (8.78) 2616.00 0.073
Vascular Age
Discrepancy
12.09 (10.04) 11.40 (10.55) 13.07 (9.24) 2746.00 0.181
BMI 28.71 (6.16) 28.53 (5.60) 29.13 (6.98) 0.361 0.549
Systolic BP 134.70
(15.33)
134.65
(15.24)
134.89
(15.50)
0.010 0.922
Diastolic BP 85.02 (12.40) 84.83 (12.02) 85.39 (12.95) 0.081 0.776
Hypertension (%) 58.4 57.3 60.6 0.177 0.674
Dyslipidemia (%) 26.7 17.7 39.4 9.433 0.002
T2DM (%) 9.9 8.4 12.1 0.630 0.427
Obesity (%) 34.8 35.8 33.3 0.104 0.748
Current Smoking (%) 23.6 25.0 21.2 0.313 0.576
Table 3. Vascular risk clinical data.
Abbreviations: HIV- = HIV-uninfected; PLWH = people living with HIV; CVD = cardiovascular
disease; BMI = body mass index; BP = blood pressure; T2DM = type 2 diabetes mellitus.

91
MRI Subset
(n = 91)
Controls
(n = 48)
PLWH
(n = 43)
Statistic P-
Value
Age 56.38 (7.92) 55.73
(8.08)
57.12
(7.77)
0.694 0.407
Gender (% Male) 73.6 70.8 76.7 0.408 0.523
Race (% Black) 67.0 68.8 65.1 0.136 0.713
Education 14.01 (2.33) 14.04
(2.34)
13.98
(2.34)
0.017 0.895
WRAT (standardized) 99.76 (15.81) 99.58
(15.15)
99.95
(16.70)
0.012 0.912
Hepatitis C (%) 5.6 2.1 9.5 2.363 0.124
Socioeconomic Status
Hollingshead Total 41.18 (12.45) 39.71
(12.90)
42.81
(11.86)
1.419 0.237
Household Income* 32.63 (34.77) 43.14
(38.57)
21.87
(26.32)
436.00 <0.001
Income Per Family* Member 27.01 (24.11) 33.12
(21.56)
20.28
(25.22)
455.00 <0.001
Psychiatric/Stressors
BDI-2 6.14 (6.24) 5.46 (6.05) 6.91
(6.43)
1.225 0.271
Current Depression (%) 2.2 4.2 0.0 1.832 0.176
Past Depression (%) 15.4 14.6 16.3 0.050 0.823
Past Mania or Psychosis (%) 4.4 0.0 9.3 4.670 0.031
CCSE 36.76 (23.57) 35.81
(8.90)
37.81
(23.88)
979.00 0.673
Health-related Stress 9.38 (8.35) 7.46 (7.58) 11.53
(8.73)
732.00 0.017
Substance Use
Past Substance Dependence
(%)
35.2 16.7 57.1 16.016 <0.001
Current Marijuana/Alcohol
Abuse (%)
11.0 12.5 9.5 0.201 0.654
Table 4. Demographic and clinical characteristics of participants who completed MRI
evaluation.
Abbreviations: WRAT (Word Reading Association Test); SES (socioeconomic status); BDI =
Beck Depression Inventory; CCSE = cumulative chronic stressor exposure; VL = viral load;
cART = combined antiretroviral therapy
*Values divided by 1000

92
Follow-Up (n
= 79)
Controls
(n = 40)
PLWH
(n = 39)
Statistic P-
Value
Age 56.70 (7.59) 55.90
(7.89)
57.51
(7.28)
0.890 0.348
Gender (% Male) 72.2 70.0 74.4 0.187 0.666
Race (% Black) 67.1 67.5 66.7 0.006 0.937
Education 13.85 (2.34) 14.05
(2.21)
13.64
(2.49)
0.599 0.441
WRAT (standardized) 99.06 (18.13) 100.70
(17.28)
97.38
(19.05)
0.657 0.420
Hepatitis C (%) 7.6 2.5 12.8 3.117 0.077
Socioeconomic Status
Hollingshead Total 39.70 (12.67) 39.15
(13.03)
40.27
(12.45)
0.152 0.697
Household Income* 32.00 (27.88) 38.97
(22.73)
25.93
(31.06)
328.00 <0.001
Income Per Family* Member 27.43 (23.59) 32.97
(19.25)
21.88
(26.36)
341.50 0.001
Psychiatric/Stressors
BDI-2 6.19 (6.73) 4.50 (5.66) 7.92
(7.34)
552.50 0.024
Current Depression (%) 2.5 2.5 2.6 0.000 0.986
Past Depression (%) 17.7 12.5 23.1 1.515 0.218
Past Mania or Psychosis (%) 3.8 0.0 7.7 3.198 0.074
CCSE 34.04 (24.90) 33.68
(24.53)
34.41
(25.59)
773.00 0.945
Health-related Stress 9.58 (8.65) 8.10 (7.62) 11.10
(9.45)
634.00 0.151
Substance Use
Past Substance Dependence
(%)
34.2 17.5 51.3 10.627 0.001
Current Marijuana/Alcohol
Abuse (%)
6.3 5.0 7.7 0.272 0.602
Table 5. Demographic and clinical characteristics of participants who completed follow-up
visit.

93
Regional ROI Individual ROI B (SE) T-statistic P-value
ACC - -6.21 (4.00) -1.554 0.124
Rostral ACC -4.92 (5.08) -0.968 0.336
Caudal ACC -7.51 (4.52) -1.660 0.101
OFC - -14.49 (5.32) -2.727 0.008^
Lateral OFC -13.46 (7.17) -1.878 0.064
Medial OFC -15.53 (6.32) -2.457 0.016
dlPFC - -33.28 (14.63) -2.274 0.026
Superior frontal -49.33 (25.01) -1.973 0.052
Caudal middle
frontal
-14.93 (11.52) -1.296 0.199
Rostral middle
frontal
-35.59 (22.82) -1.599 0.123
vlPFC - -7.35 (4.37) -1.681 0.097
Pars opercularis -10.26 (8.30) -1.236 0.220
Pars triangularis -10.14 (6.40) -1.584 0.117
Pars orbitalis -1.65 (3.98) -0.415 0.679
Table 6. Relationships between CCSE and brain volumes in the PFC.
Abbreviations: ROI = region of interest; B = unstandardized regression coefficient; SE =
standard error; ACC = anterior cingulate cortex; OFC = orbitofrontal cortex; lOFC = lateral
OFC; mOFC = medial OFC; dlPFC = dorsolateral PFC; vlPFC = ventrolateral PFC
^Significant after FDR correction for ACC, OFC, dlPFC, and vlPFC regions.

94

Figure 1. Conceptual Diagram of Primary Study Aims.

95

Figure 2. Relationship between CCSE and Global Cognition. CCSE was inversely associated
with global cognition (β = -0.395), p = 0.003). Abbreviations: CCSE = cumulative chronic stress
exposure.

96

Figure 3. Relationship between CCSE and processing speed. CCSE was inversely associated
with processing speed (β = -0.395), p = 0.003). Abbreviations: CCSE = cumulative chronic stress
exposure.

97

Figure 4. Relationship between CCSE and executive functioning. CCSE was inversely
associated with executive functioning (β = -0.253), p = 0.038). Abbreviations: CCSE =
cumulative chronic stress exposure.

98

Figure 5. Relationship between CCSE and PFC volume. CCSE was negatively associated
with PFC volume (β = -0.299, p = 0.007). Abbreviations: CCSE = cumulative chronic stress
exposure; PFC = prefrontal cortex; mm
3
= multimeter cubed.

(mm
3
)
99

Figure 6. Relationship between CCSE and OFC volume. CCSE was negatively associated
with PFC volume (β = -0.299, p = 0.007). Abbreviations: CCSE = cumulative chronic stress
exposure; OFC = orbitofrontal cortex; mm
3
= multimeter cubed.

(mm
3
)
100

Figure 6. Relationship between CCSE and dlPFC volume. CCSE was negatively associated
with dlPFC volume (β = -0.276, p = 0.026). Abbreviations: CCSE = cumulative chronic stress
exposure; dlPFC = dorsolateral prefrontal cortex; mm
3
= multimeter cubed.

101

Figure 8. PFC Volume Mediates Relationship between CCSE and Processing Speed.
Simplified path model to assess indirect effect of CCSE on processing speed through PFC
volume. The path coefficients (a, b, c, c’) are shown as standardized Beta coefficients. c’
represents direct effect of CCSE on processing speed. Abbreviations: CCSE = cumulative
chronic stressor exposure; PFC = prefrontal cortex. *p < 0.05

102
No MRI Data
(n = 70)
MRI Subset
(n = 91)
Statistic P-Value
Age 57.70 (6.57) 56.38 (7.92) 1.261 0.263
Gender (% Male) 62.9 73.6 2.143 0.143
Race (% Black) 58.6 67.0 1.220 0.269
HIV Status (% PLWH) 32.9 47.3 3.390 0.066
Education 13.80 (2.28) 14.01 (2.33) 0.331 0.566
WRAT (standardized) 97.34 (17.27) 99.76 (15.81) 0.852 0.357
Hepatitis C (%) 2.9 5.6 0.685 0.408
Socioeconomic Status
Hollingshead Total 39.72 (11.84) 41.18 (12.45) 0.564 0.454
Household Income* 39.80 (40.26) 32.63 (34.77) 2608.00 0.333
Income Per Family* Member 26.81 (22.92) 27.01 (24.11) 2859.50 0.969
Psychiatric/Stressors
BDI-2 5.50 (8.18) 6.14 (6.24) 2665.00 0.072
Current Depression (%) 2.9 2.2 0.071 0.790
Past Depression (%) 18.6 15.4 0.288 0.592
Past Mania or Psychosis (%) 4.3 4.4 0.261 0.609
CCSE 27.27 (21.04) 36.76 (23.57) 2395.00 0.007
Health-related Stress 8.36 (7.78) 9.38 (8.35) 2995.00 0.516
Substance Use
Past Substance Dependence (%) 28.6 35.2 0.876 0.349
Current Marijuana/Alcohol Abuse (%) 2.9 11.0 3.867 0.049
Supplemental Table 1. Comparison of baseline and MRI-subset samples.
Abbreviations: WRAT (Word Reading Association Test); SES (socioeconomic status); BDI =
Beck Depression Inventory; CCSE = cumulative chronic stressor exposure; VL = viral load;
cART = combined antiretroviral therapy
*Values divided by 1000

103
No Follow-Up
(n=82)
Follow-Up
(n=79)
Statistic P-
Value
Age 57.1 (7.19) 56.70 (7.59) 0.192 0.662
Gender (% Male) 65.9 72.2 0.745 0.388
Race (% Black) 59.8 67.1 0.932 0.334
HIV status (% PLWH) 32.9 49.4 4.496 0.034
Education, years 13.99 (2.27) 13.85 (2.34) 0.147 0.702
WRAT (standard score) 98.37 (14.75) 99.06 (18.13) 0.072 0.789
Hepatitis C (%) 1.2 7.6 4.003 0.045
Socioeconomic Status
Hollingshead Total 41.35 (11.68) 39.70 (12.67) 0.740 0.391
Household Income* 39.47 (44.27) 32.00 (27.88) 1.334 0.250
Income Per Family Member* 26.46 (23.54) 27.43 (23.59) 0.064 0.801
Psychiatric/Stressors
BDI-2 5.55 (7.53) 6.19 (6.73) 0.324 0.570
Current Depression (%) 2.4 2.5 0.001 0.970
Past Depression (%) 15.9 17.7 0.101 0.751
Past Mania or Psychosis (%) 4.9 3.8 0.002 0.963
CCSE 31.28 (20.93) 34.04 (24.90) 3116.50 0.679
Health-related stress 8.32 (7.53) 9.58 (8.65) 2995.00 0.408
Substance Use
Past Substance Dependence (%) 30.5 34.2 0.310 0.577
Current Marijuana/Alcohol Abuse
(%)
8.5 6.3 0.261 0.610
Supplemental Table 2. Comparisons of baseline and follow-up samples
Abbreviations: WRAT (Word Reading Association Test); BDI = Beck Depression Inventory;
CCSE = cumulative chronic stressor exposure
*Values divided by 1000

104
Regional
ROI
Individual
ROI Hemisphere
b-
coefficient T-statistic P-Value
ACC Caudal ACC Bilateral
-0.283
-1.660 0.101
L
-0.293
-1.655 0.102
R
-0.130
-0.700 0.486
Rostral ACC Bilateral
-0.155
-0.968 0.336
L
-0.174
-1.039 0.302
R
-0.094
-0.538 0.592
OFC Lateral OFC Bilateral
-0.240
-1.878 0.064
L
-0.126
-0.889 0.377
R
-0.341
-2.552 0.013
Medial OFC Bilateral
-0.366
-2.457 0.016
L
-0.230
-1.467 0.146
R
-0.420
-2.520 0.014
dlPFC Superior Frontal Bilateral
-0.278
-1.973 0.052
L
-0.268
-1.848 0.068
R
-0.258
-1.720 0.090
Caudal Middle Frontal Bilateral
-0.207
-1.296 0.199
L
-0.085
-0.509 0.612
R
-0.267
-1.577 0.119
Rostral Middle Frontal Bilateral
-0.211
-1.599 0.123
L
-0.058
-0.418 0.677
R
-0.340
-2.305 0.024
vlPFC Pars opercularis Bilateral
-0.186
-1.236 0.220
L
-0.204
-1.246 0.216
R
-0.111
-0.683 0.496
Pars orbitalis Bilateral
-0.072
-0.415 0.679
L
0.013
0.073 0.942
R
-0.118
-0.660 0.511
Pars triangularis Bilateral
-0.248
-1.584 0.117
L
-0.318
-2.051 0.044
R
-0.132
-0.745 0.458
Subcortical/ Hippocampus Bilateral
0.017
0.107 0.915
Insula

L
0.097
0.603 0.548
R
-0.060
-0.373 0.710
Amygdala Bilateral
-0.048
-0.311 0.757

L
0.035
0.217 0.829
R
-0.137
-0.848 0.399
Insula Bilateral
-0.058
-0.438 0.662

L
-0.043
-0.309 0.758
105
R
-0.066
-0.479 0.633
Supplemental Table 3. Relationships between CCSE and individual ROIs.
Abbreviations: ROI = region of interest; L = left; R = right; PFC = prefrontal cortex; ACC =
anterior cingulate cortex; OFC = orbitofrontal cortex; dlPFC = dorsolateral PFC; vlPFC =
ventrolateral PFC

106
Stressor
Index
Global
Cognition
Processing
Speed
Executive
Functioning Learning Memory
PFC
Volume
Total
Stressor
Count -0.169 -0.253* -0.147 -0.077 -0.048 -0.185
Total
Stressor
Severity -0.284* -0.383** -0.249 -0.136 -0.133 -0.232*
Chronic
Stressor
Count -0.210 -0.242* -0.188 -0.110 -0.092 -0.251*
Chronic
Stressor
Severity -0.288* -0.395** -0.253* -0.134 -0.115 -0.299**
Acute
Stressor
Count -0.117 -0.209 -0.096 -0.055 -0.027 -0.125
Acute
Stressor
Severity -0.201 -0.268* -0.168 -0.109 -0.123 -0.108
Supplemental Table 4. Associations of different lifetime stress indices with cognition and
PFC volume. Standardized regression coefficients (b) shown. *p < 0.05 **p < 0.01
Note: Chronic Stressor Severity = CCSE. Abbreviations: PFC = prefrontal cortex.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Untreated hyperglycemia associated with tau pathology and worse cognitive performance in older adults

PDF

Vascular contributions to brain aging along the Alzheimer's disease continuum

PDF

Heart, brain, and breath: studies on the neuromodulation of interoceptive systems

PDF

Cross-sectional association of blood pressure, antihypertensive medications, MRI volumetric measures and cognitive function scores in an aging population

PDF

Mechanisms of stress effects on learning and decision making in younger and older adults

PDF

A virtual reality exergaming system to enhance brain health in older adults at risk for Alzheimer’s disease

PDF

Psychosocial contributions to hippocampal volume and memory

PDF

Relationship between brain structure and motor behavior in chronic stroke survivors

PDF

The association of cerebrovascular disease risk factors with brain structure and its modification by genetic variation

PDF

Blood-brain barrier pathophysiology in cognitive impairment and injury

PDF

Indexing cerebrovascular resistance in cognitive decline and Alzheimer's disease

PDF

The role of vascular dysfunction in cognitive impairment

PDF

The acute impact of glucose and sucralose on food decisions and brain responses to visual food cues

PDF

The role of blood pressure variability in cognitive decline, cerebrovascular disease and Alzheimer’s disease

PDF

Exploring the gut-brain axis in Parkinson’s disease and the influence of physical fitness to restore gut health

PDF

Renin-angiotensin system antihypertensive medicines with and without blood-brain-barrier crossing potential: effects on cognitive outcomes in the elderly

PDF

The protective role of progenitor cells: cell proliferation relates to cognition and white matter microstructure

PDF

Associations of ambient air pollution exposures with perceived stress in the MADRES cohort

PDF

Pharmacogenetic association studies and the impact of population substructure in the women's interagency HIV study

PDF

3D vessel mapping techniques for retina and brain as an early imaging biomarker for small vessel diseases

Asset Metadata

Creator McIntosh, Elissa Charney (author)

Core Title Relationships between lifetime chronic stress exposure, vascular risk, cognition, and brain structure in HIV

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Degree Conferral Date 2021-08

Publication Date 07/24/2021

Defense Date 05/12/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag brain structure,cardiovascular disease,chronic stress,cognitive,HIV,MRI,neuroimaging,neuropsychology,OAI-PMH Harvest,vascular

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thames, April (committee chair), Bechara, Antoine (committee member), John, Richard (committee member), Mather, Mara (committee member), Meyerowitz, Beth (committee member)

Creator Email ecmcinto@usc.edu,mcintosh.elissa@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15621176

Unique identifier UC15621176

Legacy Identifier etd-McIntoshEl-9863

Document Type Dissertation

Format application/pdf (imt)

Rights McIntosh, Elissa Charney

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu