University of Southern California Dissertations and Theses

Social engagement and cognitive decline in Mexican Americans: implications for age-friendly cities

(USC Thesis Other)

Social engagement and cognitive decline in Mexican Americans: implications for age-friendly cities

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

SOCIAL ENGAGEMENT AND COGNITIVE DECLINE IN MEXICAN AMERICANS:
IMPLICATIONS FOR AGE-FRIENDLY CITIES

by

Iris Aguilar

A Dissertation Presented to the
FACULTY OF THE USC SOL PRICE SCHOOL OF PUBLIC POLICY
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF POLICY, PLANNING, AND DEVELOPMENT

August 2021

Copyright 2021 Iris Aguilar

i

COMMITTEE IN CHARGE OF CANDIDACY
Chair: Emma Aguila, PhD
Associate Professor
Sol Price School of Public Policy
University of Southern California

Kathleen H. Wilber, PhD
Mary Pickford Foundation Professor of Gerontology
Professor, Health Services Administration
Leonard Davis School of Gerontology
Sol Price School of Public Policy
University of Southern California

Kyriakos S. Markides, PhD
Annie and John Gnitzinger Distinguished Professor of Aging
Professor, Department of Preventive Medicine and Population Health
Editor, Journal of Aging and Health
Director, Texas Resource Center Minority Aging Research
University of Texas Medical Branch

Laura Trejo, MSG, MPA
General Manager
City of Los Angeles Department of Aging

ii

DEDICATION
I dedicate this work to my parents, Luis and Malvina; my sons, Daniel and Alexander;
and my sister, Hania, and nephew, Donovan. Gracias, mamá y papá por siempre alentarme y
darme el apoyo necesario para que este sueño fuese realidad. To my boys, Daniel and
Alexander, I hope you are inspired to set goals and work hard to achieve them.
I share this work in memory of Hania Irica, my sister, my friend, and my confidant,
whose spirit was with me every day as I compartmentalized my life to be able to finish this
dissertation. Se que desde el cielo estas celebrando con nosotros.

iii

ACKNOWLEDGEMENTS
First, I wish to thank May Ma Ross for walking into my office one day to encourage me
to look into the DPPD program. This journey would not have been possible without the informal
mentorship of Dr. William A. Vega, who helped me refocus my topic, calmed me when I
panicked, and always made me laugh. I am grateful for the friendship of Dr. Hector M.
González, who was my cheerleader and therapist while subtly giving me dissertation advice. To
my compadres John and Lorena Castellanos, a special thanks for stepping in to help with my
boys as I jumped into the life of a doctoral student. A warm thank you to my comadres Estela
Leon, Maria Elena Hernandez, and dear friend Mery Guerra, whose friendship gave me the
mental health support needed to complete this journey.
My sincere gratitude to my chair, Emma Aguila, who last year accepted the role and
advanced me to the finish line. Kate Wilber, you truly have been more than an informal co-chair,
helping me through the entire process. You welcomed me into your lab—where I was inspired
and energized to do research. Laura Trejo, thank you for decades of being a mentor and helping
me focus the policy implications. I am thankful to Kokos Markides, who without a second
thought joined my committee to share his vast knowledge of the data. Thank you, Emma, Kate,
Laura, and Kokos for being generous with your time and honestly being the best dissertation
committee. Thank you, Dr. María Aranda, Dr. Jorge Peniche, Dr. Donald Lloyd, Dr. Nasim
Ferdows, and Dr. Phillip Cantu for being a phone call away when I panicked over the data
analysis. I would be remiss if I did not acknowledge and thank Dr. Pamela McCann, who helped
me get through a very personally challenging semester. Last, I give a nod to the collaborative
environment of my DPPD cohort, especially Coral Andrews and Bijan Karimi. To my village,
¡Mil Gracias!
iv

TABLE OF CONTENTS
Dedication ....................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Tables ................................................................................................................................ vii
List of Figures .............................................................................................................................. viii
Abstract .......................................................................................................................................... ix
Chapter 1: Social Engagement and Cognitive Decline in Mexican Americans: Implications for
Age-Friendly Cities ......................................................................................................................... 1
Introduction ............................................................................................................................... 1
Cognitive Decline: An Impending Public Health Crisis ...................................................... 3
Social Determinants of Health ............................................................................................. 5
Protective Cognitive Factors ............................................................................................... 5
Age-Friendly Concept ......................................................................................................... 7
Purpose Statement ..................................................................................................................... 8
Research Question .................................................................................................................... 8
Conceptual Models.................................................................................................................... 9
Research Design Overview ....................................................................................................... 9
A Gap in Knowledge Remains for an Exploding Demographic ............................................. 10
Chapter 2: Literature Review ........................................................................................................ 12
Stages in Late Life .................................................................................................................. 12
Social Determinants of Health ................................................................................................ 13
Cognitive Impairment and Dementia ...................................................................................... 14
Modifiable Risk Factors .................................................................................................... 17
v

Social Engagement as a Protective Factor ......................................................................... 19
Relevant Cognition and Social Engagement Research ...................................................... 21
Theoretical Frameworks Linking Social Engagement and Cognitive Impairment ................ 26
Frameworks to Support an Aging Society .............................................................................. 29
Chapter 3: Methodology ............................................................................................................... 33
Sample and Population ........................................................................................................... 33
Instrumentation ....................................................................................................................... 35
Selected Variables ............................................................................................................. 35
Data Analysis .......................................................................................................................... 38
Institutional Review Board ............................................................................................... 41
Chapter 4: Data Reporting and Analysis ...................................................................................... 42
Descriptive Statistics ............................................................................................................... 42
Correlation Analysis ............................................................................................................... 44
Logistic Regression: Odds Ratios ........................................................................................... 46
Chapter 5: Discussion and Policy Recommendations .................................................................. 49
Study Limitations .................................................................................................................... 53
Policy Implications ................................................................................................................. 54
Transforming Health in All Policies to Social in All Policies .......................................... 54
Change or Loss of Social Network .................................................................................... 55
Health Care System Considerations .................................................................................. 56
Adverse Location Settings ................................................................................................ 57
The Role of Age-Friendly Cities in Supporting Change .................................................. 58
Conclusion .............................................................................................................................. 59
vi

References ..................................................................................................................................... 61
Appendices .................................................................................................................................... 83
Appendix A. Institutional Review Board Approval ............................................................... 83
Appendix B. STATA Commands ........................................................................................... 84
Appendix C. Logistic Regressions .......................................................................................... 89

vii

LIST OF TABLES
Table 1. Population by Age Group: Projections from 2020 to 2060 .............................................. 2
Table 2. Selected H-EPESE Domains .......................................................................................... 10
Table 3. Features of Social Engagement ....................................................................................... 20
Table 4. Cross-Referencing Age-Friendly Domains and Dementia-Friendly Sectors ................. 31
Table 5. H-EPESE Sample Size by Data Collection Year ............................................................ 34
Table 6. List of Social Engagement Variables .............................................................................. 36
Table 7. List of Health Behavior Variables .................................................................................. 37
Table 8. List of Health Condition Variables ................................................................................. 38
Table 9. Descriptive Statistics ....................................................................................................... 42
Table 10. Correlations between Cognition and Covariates .......................................................... 45
Table 11. Odds Ratio Summary: Men, Women, and Full Sample ............................................... 46
Table 12. Examples of Potential Triggers Interfering with Social Engagement .......................... 55

viii

LIST OF FIGURES
Figure 1. Projected Number of Children and Older Adults ............................................................ 2
Figure 2. Racial Disparities in Future Burden of Dementias .......................................................... 3
Figure 3. Social Determinants of Health ......................................................................................... 5
Figure 4. Alzheimer’s Disease Continuum ................................................................................... 15
Figure 5. Modified NASEM Guiding Framework ........................................................................ 28
Figure 6. Social Trajectory Model ................................................................................................ 29
Figure 7. Flowchart of Sample Selection ...................................................................................... 40

ix

ABSTRACT
Forecasts of the older adult population indicate that the 85 and older group is expected to
increase more than two-fold, with a 123% increase, by 2060. Along with longer life expectancy,
there is increased diversity in all racial and ethnic groups. The overall Hispanic and Latino
(hereafter Latino) population is expected to outpace all other racial and ethnic groups, growing
from 18% in 2016 to 22% in 2035. Latinos live longer (2.5 years), but they do so with higher
levels of multiple chronic health conditions, disability, and cognitive impairment. Existing
studies suggest social engagement is a dementia protective factor. To answer the research
question of what features of social engagement are associated with lower severe cognitive
decline, binary logistic regressions were performed using the longitudinal Hispanic Established
Populations for the Epidemiologic Study of the Elderly data collected in 2004–05 and 2006–07.
Participants at baseline with severe cognitive decline were excluded from the analysis to avoid
reverse causality. After participants were classified as having normal cognition or likely
dementia based on the 6-year follow-up data, the final analytic sample was N = 863. The odds
ratio indicates that likely dementia decreases with the frequency of contact with friends,
perceived support from family members, marriage (for women), and higher education levels.
Policymakers and practitioners can influence population health through multisystem mechanisms
using a “social in all policies” approach. The proactive strategies of social in all policies will
create age-friendly environments that allow individuals of all ages to thrive.
1

CHAPTER 1: SOCIAL ENGAGEMENT AND COGNITIVE DECLINE IN MEXICAN AMERICANS:
IMPLICATIONS FOR AGE-FRIENDLY CITIES
Introduction
The population in the United States is increasingly older due to the aging of the baby
boomer generation—born post-World War II (U.S. Census Bureau, 2017). Researchers have also
attributed these demographic changes to lower fertility and increased life expectancy largely
related to biomedical innovations and immigration trends (Lichtenberg, 2017; Mizoguchi et al.,
2019; Roberts et al., 2016; World Health Organization [WHO], 2015). A population trend has
begun an unprecedented shift toward an older society. In the United States, people aged 65 or
older outnumber the number of children aged 5 or younger (U.S. Census Bureau, 2019). The
baby boomers are projected to continue to transform the age structure as they move into the
oldest-old age group—aged 85 or older (Ortman et al., 2014). Figure 1 shows the projected
population numbers for children and older adults through 2060 as reported by U.S. Census
Bureau in a report titled An Aging Nation (Ortman et al., 2014). After considering fertility for
native- and foreign-born women, mortality accounting for the higher life expectancy of foreign-
born persons, and immigration patterns, the report projected a demographic shift in 2035, when 1
in 5 persons will be older than 65 (Vespa et al., 2018).
2

Figure 1. Projected Number of Children and Older Adults

It is expected that population changes will also occur in the older adult population. As
shown in Table 1, the population by age group projections for the 85 or older group is expected
to increase more than two-fold, with a 123% increase, by 2060 (Vespa et al., 2018). The increase
of people living well into their 80s has led to the creation of subcategories to better describe their
life stage: ages 65–74 as young-old; 75–84 as middle-old; and 85+ as oldest-old adults
(Settersten & Mayer, 1997).
Table 1. Population by Age Group: Projections from 2020 to 2060

According to the U.S. Census Bureau (2017), along with longer life expectancy, we are
experiencing increased diversity in all racial and ethnic groups. The overall Hispanic and Latino
(hereafter Latino) population is expected to outpace all other racial and ethnic groups, growing
from 18% in 2016 to 22% in 2035 (U.S. Census Bureau, 2018).
2016 2020 2030 2040 2050 2060
Total 100.00 100.00 100.00 100.00 100.00 100.00
.Under 18 years 22.79 22.24 21.30 20.65 20.11 19.81
.18 to 64 years 61.97 60.91 58.10 57.71 57.86 56.78
.65 years and over 15.24 16.85 20.60 21.64 22.03 23.41
.85 years and over 1.97 2.01 2.56 3.86 4.77 4.70
.100 years and over 0.03 0.03 0.04 0.05 0.10 0.15
3

Cognitive Decline: An Impending Public Health Crisis
The Centers for Disease Control and Prevention has stated that Alzheimer’s disease, the
most common type of dementia, is an irreversible debilitating condition linked to progressive
memory loss and impairment of other cognitive functions—known as cognitive impairment. The
term dementia refers to a syndrome that includes symptoms that impair memory and other
thought processes and interfere with everyday activities (Lezak, 1995). The most significant risk
factor for Alzheimer’s disease and related dementias (hereafter dementia) is age (Alzheimer’s
Association, 2019). In addition to age, racial and ethnic disparities have been attributed to the
development of dementia, such as longevity, obesity, cardiovascular risk, diabetes, and other
socioeconomic determinants such as education (Ai et al., 2011; National Center for Health
Statistics, 2019). Figure 2 shows the estimated racial disparities in future dementia cases through
2060 (Alzheimer’s Association, 2021).
Figure 2. Racial Disparities in Future Burden of Dementias
Medicare Beneficiaries: Aged 65 and Older with Alzheimer’s Disease by Race and Ethnicity
13.8%
African American

12.2%
Hispanics
10.3%
Non-Hispanic Whites
Alzheimer’s disease projected to increase 16% by 2060

6.2 million
2021
13.8 million
2060
It is critical to understand that dementia falls along a continuum of cognitive decline that
has a preclinical phase (no ascertainable changes) and a clinical phase that starts with mild
cognitive impairment, then mild dementia followed by moderate dementia and finally severe
dementia (Aisen et al., 2017; Alzheimer’s Association, 2020; Campbell et al., 2013). People with
mild cognitive impairment have been found to progress to dementia “at a rate of 3 to 5 times
4

higher than those with normal cognition” (Campbell et al., 2013). The progressive feature of
dementia provides an opportunity to intervene and change the course of the disease (Aisen et al.,
2017; Alzheimer’s Association, 2020).
Although dementia can affect anyone, some groups, including Latinos, are at higher risk
(Gaugler et al., 2019). It has been well documented that on average compared to the population,
Latinos live longer (2.5 years), but they do so with higher levels of multiple chronic health
conditions, disability, and cognitive impairment (R. J. Angel et al., 2015; Markides & Eschbach,
2011; W. A. Vega et al., 2015). This study focused on Latinos given their higher risk of
developing dementia and projected demographic growth, along with a continued lack of
inclusion in research (Gonzalez, 2020).
The most recent Centers for Disease Control and Prevention report on mortality listed
dementia as the sixth leading cause of death in the United States and as the third for California
(National Center for Health Statistics, 2018). Unless a cure for Alzheimer’s disease is found, we
can expect the projected trends of dementia to have devasting impacts on families and financial
systems of care. The cumulative cost of medical and long-term care expenditures along with
indirect care expenses such as informal care (based in 2012 dollars) will cost the U.S. economy
an estimated $2.35 trillion by 2060 (Wu et al., 2016). For Latinos families, that tend to provide
care at home, the experience of caring for a relative with dementia creates more burden than in
other ethnic groups because of the longer time spent caregiving (Villa & Wallace, 2020). The
caregiving commitment creates personal distress, health problems, loss of wages from reduced
work hours or unemployment, food insecurity, and severe retirement insecurity (Rote et al.,
2019; Villa & Wallace, 2020; Wu et al., 2016).
5

Social Determinants of Health
Health behaviors and health care treatment greatly vary among cultural groups and by
socioeconomic status. A widely accepted framework for understanding these factors is known as
the social determinants of health, shown in Figure 3 (Artiga & Hinton, 2019). Social
determinants of health are defined as the conditions in which people are born, grow, work, live,
and age and the broader set of forces and systems such as economic and social policies,
developmental agendas, social norms, and political systems that shape the conditions of daily life
(Dahlgren & Whitehead, 1991; WHO, 2019). Unfortunately, social relations have received little
focus from U.S. public policies and programs (Holt-Lunstad et al., 2017). More recently,
awareness has grown regarding the importance of social factors to health outcomes (Artiga &
Hinton, 2019).
Figure 3. Social Determinants of Health
Protective Cognitive Factors
Research suggests that Latinos, particularly Mexican Americans, the largest group in the
United States, experience relatively healthy lives until middle age, when their health status
begins to decay, and old age, which is plagued with health problems and disability (González et
6

al., 2009). Recent findings from the Framingham study indicate the overall reduction in the
incidence of dementia among the non-Hispanic White population is linked to a decrease in risk
factors such as low education, smoking, and cardiovascular disease (Satizabal et al., 2016). Some
of the risk factors such as obesity, cardiovascular risk, diabetes, and other socioeconomic
determinants are considered modifiable (Livingston et al., 2020). Researchers have also
identified dementia protective lifestyle factors such as physical activity, reduction and
management of cardiovascular disease risk factors (including hypertension, diabetes, and
smoking), communication with health care providers, and social and intellectual engagement
(Blazer et al., 2015). A global call to action by the Lancet Commission to promote well-being
and prevent dementia contends that there is enough scientific evidence to accept social contact as
a dementia protective factor (Livingston et al., 2020; Livingston et al., 2017).
Although studies that include significant sample sizes of minority populations such as
Mexican Americans and other Latinos are limited (Gonzalez, 2020), existing studies suggest
social engagement is a dementia protective factor (Alzheimer’s Association, 2019; Livingston et
al., 2020; WHO, 2012). Given the increased incidence of dementia among Latinos including
Mexican Americans, understanding how features of social engagement operate is a necessary
step to addressing racial and ethnic disparities.
Social engagement is defined as having many social connections with a great deal of
participation in social activities (Bassuk et al., 1999). Berkman and Glass (2000) recognized that
social participation and social engagement is a “difficult to define pathway by which [social]
networks may influence health status.” Nonetheless, social networks and social connectedness
have been identified as providing protective effects against dementia (Amieva et al., 2010;
Baumgart et al., 2015; Pillai & Verghese, 2009; Strout & Howard, 2012; Yaffe, 2018;
7

Zunzunegui et al., 2003). Social engagement can be described as activities such as church
attendance, getting together with friends, attending social functions, group recreation, and
participating in occupational or social roles (Berkman & Glass, 2000; Glass, 2000).
The Global Council on Brain Health (GCBH), an independent group of scientists, health
professionals, scholars, and policy experts convened by AARP from around the world to tackle
brain health as it relates to cognition, released a report with recommendations based on
consensus from multidisciplinary experts on social engagement and brain health. It defined
social engagement as interacting with others, feeling connected to other people, doing purposeful
activities with others, or maintaining meaningful social relationships (Carlson et al., 2017). The
report highlighted the social connection components of social engagement that influence brain
health.
Regarding social engagement, it is essential to point out that it has a cultural dimension
that in the Latino population may be related to sociodemographic variables (age, gender, and
marital status) including health status (Markides et al., 1986; Rodríguez-Galán & Falcón, 2010).
Investigations of Mexican American older adults have indicated social networks and engagement
facilitate seeking help and maintenance of healthier lifestyles (Baxter et al., 1998; Hansen &
Aranda, 2012; Hill et al., 2005). Mexican Americans, particularly older adults, are strongly
connected with their children and engage in mutual social support (J. L. Angel et al., 2004;
Markides et al., 1986; Markides et al., 2013). Further discussion of social engagement through
the Mexican American experience is included in the literature review chapter.
Age-Friendly Concept
In preparation for the unprecedented population growth among older adults, the WHO
launched in 2006 a global campaign called Age-Friendly Cities (Alley et al., 2007; Greenfield et
8

al., 2015; Plouffe & Kalache, 2010). The WHO age-friendly cities movement seeks to promote
healthy and active aging through the promotion and creation of built environments and designing
of services to be accessible and inclusive for people throughout the life span (Hepp, 2016; Law
et al., 2014).
The WHO partnered with Alzheimer’s Disease International to publish a call to action,
citing dementia as a public health priority (WHO, 2012). As a result, in 2015, the Dementia
Friendly America initiative was established in the United States as “a movement to more
effectively support and serve those across America who are living with dementia and their family
and friend care partners” (Dementia Friendly America, 2015). Across the United States, about
522 communities have joined the Age-Friendly Network (AARP Livable Communities, 2021).
Purpose Statement
The purpose of this study was to examine social engagement during late life among older
Mexican Americans not found to have severe cognitive impairment. The study explored the role
of social engagement in protecting against or delaying cognitive decline and the implications of
social engagement in age-friendly cities policies. The age-friendly cities initiative offers
opportunities to develop policies and programs that address social engagement through social
participation activities across the lifespan. Public policy can promote projects and programs that
address social and physical barriers such as intergenerational interactions, civic engagement, and
walkable neighborhoods to foment social interactions (Scharlach & Lehning, 2013).
Research Question
What features of social engagement in late life are associated with lower prevalence of
likely dementia? The goal was to examine the social engagement of respondents without severe
cognitive decline to minimize reverse causality. Based on prior studies conducted with various
9

racial and ethnic groups, my hypothesis was that social engagement in late life serves as a
protective cognitive factor.
Conceptual Models
A more detailed discussion of the conceptual models briefly described in this paragraph
can be found as part of the literature review in Chapter 2. The first model to be described is the
National Academies of Science, Engineering, and Medicine (NASEM) social isolation and
loneliness guiding framework, which conceptualizes the role of social engagement (social
connections) and its relation (causal flow) to cognitive impairment (health impacts) at the
individual level with a clear embeddedness in larger systems at the community and society
levels. A second theoretical framework providing context for this dissertation is the life course
perspective described in the social trajectory model, which defines the pathways that lead to
adult health outcomes. The social trajectory model links early life social conditions to adult
social conditions and adult health outcomes (Berkman, 2009). This model is of particular
relevance for initiatives that seek to address aging across the lifespan such as age-friendly cities.
Berkman et al. (2011) pointed out that understanding risk factors across the lifespan also requires
recognizing that differences exists between young-old (65-74) and old-old (75+) adults and that
young-old adults need help to prepare to be old-old adults—for example, with policies that foster
engagement in old age.
Research Design Overview
To answer the research question, I used data from the Hispanic Established Populations
for the Epidemiologic Study of the Elderly (H-EPESE) collected in 2004–05 and 2010–11. The
H-EPESE is a longitudinal study that started collecting data from Mexican Americans who
reside in Texas, California, New Mexico, Arizona, and Colorado in 1993–94, when participants
10

were 65 years old or older (Markides, 1993-2011). The dataset includes cross-sectional and
longitudinal information about health, living arrangements, and quality of life issues. The H-
EPESE was replenished during Wave 5 in 2004–05 with a representative sample of 902 adults
aged 75 or older (Downer, Al Snih, et al., 2019), increasing the H-EPESE sample to 2,269
Mexican Americans. To test my hypothesis, I used the selected H-EPESE variables in the
domains listed in Table 2. A full discussion of the variables is included in Chapter 3.
Table 2. Selected H-EPESE Domains
Domains Variable Type
Cognitive Status Dependent
Social Engagement Primary Independent
Health Behaviors Other Independent
Health Conditions
Sociodemographics
A Gap in Knowledge Remains for an Exploding Demographic
Several longitudinal population-based studies in the United States, Europe, and Asia have
identified social engagement as having beneficial effects on cognition in old age (for a review,
see Fratiglioni et al., 2004; Kuiper et al., 2015). In the United States, Ertel et al. (2008) found
that social integration delays memory loss using data from the Health and Retirement Study.
Seeman et al. (2011), using data from the National Midlife in the United States, reported that
social engagement is linked to cognitive function. Although these studies used nationally
representative cohorts, the samples were composed of non-Hispanic White and non-Hispanic
Black participants. Research on the association of social engagement and cognitive decline
among Latinos in the United States remains limited, especially in the Mexican American
population (Alzheimer’s Association, 2019; Amano et al., 2019; Lara et al., 2019).
A search of the literature of studies focused on Mexican Americans revealed two
longitudinal studies, Howrey et al. (2015) and Rote et al. (2021). Howrey et al. (2015) used 18
11

years of the H-EPESE data and found an association of reduced cognitive decline with church
attendance. Still, they did not find a causal relationship between social support and the onset of
cognition-related disability (Howrey et al., 2015). The social support variable was derived from
two questions on perceived support (“Can you count on at least some of your family?” and “Can
you talk about your deepest problems with at least some of your family or friends?”). The
inclusion of all respondents, including those with cognitive impairment, may have created
bidirectional causality from family or friends increasing social support as a result of cognitive
decline.
Rote et al. (2021) sought to describe long-term patterns through trajectory modeling
social support, and dementia using seven waves (1992–93 to 2010–11) of the H-EPESE. Social
support was defined as perceived emotional support with the same items used in the Howrey
study and instrumental support (“Can you count on some of your family?”). Low support was
found to be higher in persons with likely dementia. The study did not use other features of social
engagement, limiting the analysis to two questions (one perceived and one instrumental). The
analysis also included persons with cognitive impairment, called likely dementia, which has been
found to have bidirectional causality and be related to lower trajectories of social support.
This study included H-EPESE participants without severe cognitive decline or likely
dementia, using a more comprehensive model with features of social engagement to provide new
information about its relationship with cognitive function and implications for improving age-
friendly policies.

12

CHAPTER 2: LITERATURE REVIEW
This literature review is not exhaustive but rather highlights research studies that provide
a better understanding of social engagement and cognitive impairment among Latinos and
whenever possible, focused on Mexican Americans. Latinos in the United States are inclusive of
any race with origins from more than 25 countries (Alzheimer’s Association, 2004). Mexican-
origin Latinos (U.S.-born and immigrants) are the largest subgroup, representing 62% of the
nearly 62 million (19%) Latinos in the U.S. population (Noe-Bustamante et al., 2019; U.S.
Census Bureau, 2018).
Stages in Late Life
Almost 60 years ago, in 1961, theories involving social engagement emerged. Elaine
Cumming and William Henry proposed disengagement theory to describe aging as “an
inevitable, mutual withdrawal or disengagement, resulting in decreased interaction between the
aging person and the others” in their social system (Cumming & Henry, 1961). Disengagement
theory casts the aging process with a pessimistic description of everyone retiring into old age as
reclusive individuals. As a result, a competing idea was proposed in 1964 by Bernice Neugarten
called activity theory. Neugarten asserted that “late-life satisfaction depended on the active
maintenance of one’s relationships and continual involvement in meaningful pursuits”
(Achenbaum, 2009). These theories were later critiqued for their lack of consideration of diverse
individuals from various racial and ethnic, gender, and social classes (Achenbaum, 2009;
Hochschild, 1975). Yet these seminal works have served as a springboard for many current
approaches to understanding late-life social engagement.
Neugarten’s research into late-life satisfaction led to a description of two late-life stages,
ages 65–74, called young-old adults, and ages 75 or older, called old-old adults (Neugarten,
13

1974). As she later described them, these age groups were “originally suggested as a gross way
of acknowledging some of the enormous diversity among older persons” (Neugarten, 1996). This
age group concept has evolved to include a third generational cohort as longevity increased. The
age categories for late life widely used today are 65–74 to represent young-old adults, 75–84 for
middle-old adults, and 85 or older for oldest-old adults (Settersten & Mayer, 1997). These
chronological categories acknowledge the differences between a person aged 65 and another
aged 90. The awareness of generational differences is fundamental to understanding that relevant
contextual factors and diseases are present or more common for certain age groups, including in
late life. These cohort differences are of extreme relevance because projections predict the
number of persons older than 85 will triple by 2040 (Ortman et al., 2014).
R. J. Angel et al. (2015) characterized the increased longevity as a “public health or
prevention paradox.” Using longitudinal data from the H-EPESE, they sought to determine the
proportion of serious functional impairment among Mexican Americans. Life table analyses
revealed study participants spent more than half of their life after age 65 with serious functional
limitations. In terms of health and human services policies, it is critical to understand factors that
prevent functional decline to reduce social and financial burden.
Social Determinants of Health
Using its global platform, the WHO (2019) has disseminated the concept of the social
determinants of health as the conditions in which people are born, grow, live, work, and age.
These conditions leading to mortality and morbidity are influenced by money, environment
(neighborhood and physical), education, food, health care, and social context. This dissertation
focused on the latter part that involves social context. Social context includes social integration,
support systems, community engagement, discrimination, and stress (Irwin et al., 2008; WHO,
14

2019). For older adults, social support is related to health outcomes (Office of Disease
Prevention and Health Promotion, 2014).
The research literature indicates a relationship between low socioeconomic position and
late-life dementia. Zeki Al Hazzouri et al. (2011) created indicators of socioeconomic status in
childhood, adulthood, and midlife to examine changes in socioeconomic positions across the life
course and the development of dementia among Mexican Americans (N = 1,634) in the
Sacramento Area Latino Study on Aging. Findings indicated a relationship between being
socioeconomically disadvantaged throughout life and dementia, similar to other published
studies highlighting neurodegeneration as “shaped by life-course experiences” (Zeki Al Hazzouri
et al., 2011). I. E. Vega et al. (2017) expounded on the intersection of genetics and social
determinants of health and recommended approaches that honor Latino values and social
complexities of specific populations or ethnic groups.
Cognitive Impairment and Dementia
The rapid increase in the older adult population has fostered differentiation between
normal cognitive aging and dementia. The Centers for Disease Control and Prevention has
defined cognitive impairment as “trouble remembering, learning new things, concentrating, or
making decisions” that affect daily life (Hachinski & Munoz, 2000). The interference with
everyday life is described as gradual or mild, then eventually transitioning to severe levels.
During severe cognitive impairment, the person cannot understand or communicate (Aisen et al.,
2017; Alzheimer’s Association, 2020; Hachinski & Munoz, 2000). Although cognitive
impairment has many causes, the most common in late life is Alzheimer’s disease (Alzheimer’s
Association, 2020). Alzheimer’s disease occurs along a continuum, as shown in Figure 4. The
start is a preclinical phase with no noticeable changes until it transitions into mild cognitive
15

impairment, when individuals begin to notice memory changes but can still perform their daily
activities (Alzheimer’s Association, 2020; Petersen et al., 1997).
Figure 4. Alzheimer’s Disease Continuum
Suthers et al. (2003) used a U.S. nationally representative sample of persons aged 70 or
older to estimate prevalence and years of life with and without cognitive impairment (Suthers et
al., 2003). Using data from the Assets and Health Dynamics among the Oldest Old survey, they
found that on average at age 70, most Americans will live 1.5 years with cognitive impairment.
The prevalence was found to be 9.5% for moderate to severe cognitive impairment. Women were
found to live longer with cognitive impairment due to their longer life spans. Suthers et al.
(2003) simulated life expectancy and projected that by 2040, cognitive impairment will more
than triple. This survey merged with the Health Retirement Survey in 1998 to continue to
provide data regarding community-dwelling older adults. Recent findings continue to show a
trend toward longer life expectancies with increased prevalence of impairment, particularly for
Latinas (Hale et al., 2020). Although the COVID-19 pandemic has created additional
uncertainties about the fertility, mortality, and immigration rates linked to these trends, the
challenges of cognitive impairment among the oldest-old adults remains a critical issue (E. Clark
et al., 2020; Korczyn, 2020; Shiels et al., 2021).
Brookmeyer et al. (1998) conducted a study to project the prevalence and incidence of
Alzheimer’s to discuss the potential impact of delaying the onset of symptoms. Drawing from
16

data collected in four U.S. studies, they used regression analysis methods to obtain age-specific
incidence rates multiplied by the population based on U.S. census estimates. Using a formula
that considered age and disease onset, the authors found that a 5-year delay (50% risk reduction)
would reduce prevalence by 1.15 million people after 10 years. A 6-month delay indicated
reductions of nearly 100,000 after 10 years and 380,000 after 50 years. The public health cost
savings from a 6-month delay was estimated at $4.7 billion at 10 years and nearly $18 billion
annually after 50 years (Brookmeyer et al., 1998).
The few research studies on cognitive impairment related to dementia indicate a higher
prevalence among Latinos. O’Bryant et al. (2013) analyzed data from two Alzheimer’s disease
and cognitive aging research studies to examine the risk of mild cognitive impairment. The data
included medical evaluations, neuropsychological tests, biomarkers, and standardized interviews.
Mexican Americans (n = 626) were compared to non-Hispanics (n = 1,002) using continuous
analysis of variance and logistic regression analyses. Mexican Americans were found to have a
higher incidence of health conditions, lower educational levels, and lower presence of the gene
associated with Alzheimer’s, APOE e4, despite on average being younger (68 vs. 72 years old).
Age was found to be the most significant risk factor for mild cognitive impairment among
Mexican Americans (O’Bryant et al., 2013).
Downer et al. (2018) used H-EPESE data for adults aged 75 or older from 1993 to 2016
to assess cognitive status and cognitive decline between the original H-EPESE cohort (baseline
1993–94), characterized by lower educational levels, and the cohort added in 2004–05. The study
hypothesized that the 2004–05 cohort would have higher odds of cognitive decline but slower
decline and lower incidence rates. To test the hypothesis, a subsample of respondents with direct
interview data on dependent variable (cognitive status) and covariates (age, sex, education,
17

nativity, diabetes, heart disease, stroke, and depression) was selected. Cognitive function was
determined using normalized Mini-Mental State Exam (MMSE) scores based on latent class
models developed by Phillips et al. (2014) to minimize inaccurate scores for those with very high
or very low cognition. The analysis modeled prevalence of cognitive impairment, trajectories of
cognitive decline, and incidence of cognitive impairment. The 2004–05 cohort, aged 75 or older,
was found to have 2.5 times higher risk of being cognitive impaired than their peers in the 1993–
94 cohort. The prevalence of cognitive impairment was elevated for 2004–05 cohort respondents
with less than 4 years of education. This cohort also had slower rates of cognitive decline and
lower risk during a 10-year span (Downer, Garcia, et al., 2019). These differences suggest, at
least among H-EPESE older adults, the possibility of an effect related to a modifiable risk factor,
education. This study highlights the importance of education in cognitive research, even a 1-year
difference, and gaps in the understanding of how best to support older adults during late life,
particularly Latinos, who are projected to have the highest life expectancy by 2060 (Medina et
al., 2020).
Modifiable Risk Factors
Research has made significant inroads in drawing a clearer picture of biological changes
at the cellular level linked to Alzheimer’s and other dementias. Still, treatment options provide
minor symptom control and do not change the eventual disease outcome. There are no
prevention treatment options, only recommendations for healthy brain activities. Proceedings
from a NASEM workshop in 2018 included a chapter by Kenneth Langa on the importance of
documenting population health patterns, trends, and heterogeneity among both the immigrant
and U.S.-born segments of this rapidly growing group. Langa (2018) pointed out that much of
the dementia research involving Latino subgroups indicated differences in health patterns across
18

Hispanic subgroups, with Puerto Ricans generally exhibiting the worst population health profile,
Cubans the most positive, and the Mexican-origin population in the middle. Langa further
discussed the brain’s complexity and the multiple factors that affect the brain’s health across the
life course, acknowledging that it is doubtful a single “magic bullet” will be found that prevents
dementia in the growing population of older adults.
Langa (2018) presented information regarding the relationship between low
socioeconomic status and racial and ethnic disparities for cognitive decline and dementia in the
United States. This relationship was identified as pernicious and threatening to an aging society
that hopes to provide some equity of access to the resources and environments that allow people
to live long and well. Langa called for movement toward increased equity in access to resources
and environments that support living long and well. The author pointed out the complexity of
dementia disparities that require lifestyle factor changes embedded in cultural beliefs and values.
A more recent study examined cognitive stimulation using a heterogenous group of
Latinos participants from the Hispanic Community Health Study/Study of Latinos. The data
came from adults aged 45 or older (N = 9,438) of Central American, Cuban, Dominican,
Mexican, Puerto Rican, and South American backgrounds. Vásquez et al. (2019) created an
index for cognitive stimulation composed of education, occupation, social network, and
acculturation to examine the association with cognitive function. Their analysis indicated that
mental stimulation provides cognitive protection. They found social networks in the family,
bicultural social activities, occupation, and education may give cognitive protection for high-risk
individuals (Vásquez et al., 2019).
The Lancet Commission recently released an update to its 2017 report on dementia
prevention, intervention, and care. The 2017 report involved a global systematic review and
19

meta-analysis of research that described a life course model of potential risk factors that
individuals can modify to have an effect on the development of dementia, describing social
engagement as “a necessary condition for wellbeing throughout life” (Livingston et al., 2017). A
2020 report increased the modifiable risk factors from nine to 12, adding excessive alcohol, brain
injury, and air pollution to the previous modifiable risk factors of education, high blood pressure,
hearing impairment, smoking, obesity, depression, lack of physical activity, diabetes, and lack of
social contact (Livingston et al., 2020). The Lancet Commission found that frequent social
contact allows the maintenance of and increase in the brain’s cognitive reserve capacity.
Cognitive reserve refers to the ability of some people to continue to perform tasks despite their
brain changes, whereas others don’t have the same resilience in the face of the pathological brain
changes (Stern et al., 2020). In its review of the research literature, the Lancet Commission
found that marital status is an important contributor to social engagement and that late-middle-
age social contact seems to provide a reduction in dementia risk. In practice, the findings of the
commission emphasize opportunities to impact social engagement at the individual level but also
at the community and societal levels throughout the life span.
Social Engagement as a Protective Factor
The definition used in this dissertation for social engagement is derived from a report of
the AARP collaborative, the GCBH, composed of scientists, health professionals, scholars, and
policy experts. The GCBH reviewed current evidence and expert consensus reports to discuss
practical ways to foster brain health using reliable and scientifically based information (GCBH,
2020). Therefore, social engagement is characterized as the “social interactions that are pleasing
and meaningful to those who engage in them and have positive outcomes such as providing
emotional or practical support” (Carlson et al., 2017). The GCBH described social engagement
20

as the outcome of social connections. As shown in Table 3, Carlson et al. (2017) defined the
features of social engagement as multidimensional structural, emotional, and quality
components.
Table 3. Features of Social Engagement
Structural Components
(features of social connectedness)
Functional Components
(nature of interactions)
Quality Components
(individuals’ experience)
• Composition of group: age,
gender, cultural diversity
• Duration of contact
• Frequency of contact
• Individual vs. group activity
• Presence or absence: family
or friends, partner, spouses,
neighbors
• Size of group(s)
• Type
• Complexity (emotional and
behavioral dimensions)
• Instrumental support
• Emotional support
• Intensity
• Intergenerational dynamic
(transfer of knowledge)
• Reciprocity
• Variety
• Fun and novelty
• Joyfulness
• Meaningfulness and purposefulness
• Satisfaction with ties
• Sense of belonging
• Sense of social well-being
• Supportiveness
The ability to have a mixture of social connections creates significant features that
support cognitive function and prevent cognitive decline. The definition provided by GCBH
quilts together multiple dimensions that allow social engagement to occur.
Dimensions of Social Engagement in Mexican Americans
Latino families, including Mexican Americans, like many racial and ethnic groups, value
familismo. Familismo, or familism, refers to the importance of the family and familial
relationships in Latino culture (Grebler et al., 1970). Features of familismo include family
obligations, perceived support, and family as the role model (Sabogal et al., 1987). The family
consists of immediate members, aunts, uncles, cousins, grandparents, lifelong friends, and other
close friends. W. A. Vega (1990) found that family network sizes vary based on socioeconomic
status (low versus high). Mexican Americans tend to participate in larger kin networks and
engage in higher visiting and exchange rates than non-Hispanics. W. A. Vega (1990) pointed out
21

that this propensity leads to the family’s availability and usage of assistance or help with tangible
(instrumental) and affective needs. Family defines and shapes social engagement.
The Mexican American family tends to be characterized as patriarchal in structure, with
men and women performing traditional roles (Keefe et al., 1979; Landale et al., 2006; Markides
et al., 1986; W. A. Vega, 1990). A gender stereotyping that is malleable and dependent on
availability or need for financial resources leads to working outside of the home (W. A. Vega,
1990). Employment leads to conflicts and role strains (W. A. Vega, 1995), including changes in
social network participation (Hill et al., 2016). This study considered that gender differences
exist and based on previous research, are relevant to understanding the effects of social
engagement (Berkman & Glass, 2000; Keefe et al., 1979; Shye et al., 1995).
Relevant Cognition and Social Engagement Research
In the last two decades, research has increased on the association of cognitive decline and
social engagement or some of its features such as social support, social activities, isolation, and
social participation (Kelly et al., 2017; Kuiper et al., 2015; Penninkilampi et al., 2018). The
articles noted here include those that provided this dissertation with a general analytic
rationalization and whenever possible, reflect research specific to U.S. Latinos.
Bassuk et al. (1999) collected data from community-dwelling adults aged 65 or older (N
= 2,812) in New Haven, Connecticut, from 1982 to 1994 to examine the relationship between
social disengagement and incidence of cognitive decline. Social disengagement was
operationalized as a spouse’s presence, monthly visual contact with three or more relatives or
friends, yearly nonvisual contact with 10 or more relatives or friends, attendance at religious
services, group membership, and regular social activities. Bassuk et al. (1999) constructed a
composite index that included social activity indicators, social networks, and perceived
22

emotional support. Cognition was assessed with the Short Portable Mental Status exam assessed
four times during the 12 years. The researchers found that social disengagement was a risk factor
for cognitive impairment. The reported data tables did not indicate the ethnicity breakdown of
the non-White participants, who represented 43.0% of the sample. The composite indexes
created in this seminal study provide a framework for aggregating social engagement features.
Zunzunegui et al. (2003) examined the influence of social networks and social
engagement in a representative sample of community-dwelling Spaniards aged 65 or older (N =
964) in a suburb of Madrid, Spain. A composite social integration index was created from three
baseline variables (membership in a club, church attendance, and community center attendance).
The social engagement variable was created from social ties with children, relatives, and friends
and categorized as low, medium, or high for descriptive analysis and as a continuous variable for
multivariate analysis. It is important to note that the frequency of contact with children was
treated differently from relatives or friends and only considered an “active” social engagement
feature if it involved providing help to children. A cognition composite score was derived from
the Short Portable Mental Status questionnaire, the Barcelona Test, and the EPESE short story
recall, and persons with severe cognitive deficits were excluded to reduce reverse causality.
Using multiple linear regressions, the study found that (a) social disengagement, lack of social
activities, and poor social connections were associated with cognitive decline; (b) formal social
activities had a protective effect; and (c) there were gender differences regarding engagement
with friends, which was protective for women but for not men (Zunzunegui et al., 2003). This
study included persons with low education levels, similar to those found in the H-EPESE dataset.
Hill et al. (2006) used seven waves of the H-EPESE (N = 2,734 after proxy data were
eliminated) to examine the association between social engagement and slower rates of cognitive
23

decline. Social engagement was operationalized through a single variable, church attendance.
Cognitive function trajectories were obtained using continuous specifications of the MMSE.
Using three linear growth curve models, they found church attendance on a monthly, weekly,
and more than weekly basis was associated with slower cognitive decline rates. However,
findings also indicated patterns suggesting that poor cognitive function may limit the ability to
attend church (Hill et al., 2006). This study supports research conducted with other racial and
ethnic groups that identified church attendance as a social engagement activity.
Hughes (2013) used three waves of data from the Monongahela–Youghiogheny Healthy
Aging Team Study in Pittsburgh (N = 816) to examine the risk of progression from mild to
severe cognitive impairment when engaged in social activities. Social activity composite scores
were created from the total number of different activities (church, family and social activities,
clubs, and volunteering) and frequency of engagement. The cognition variable was derived from
the MMSE using age-education adjustment scores greater than 21 out of 30. Chi-square tests and
t-tests were used to compare social activities during each data collection wave. Using
multivariate model analyses, the study found that social engagement prevented or delayed
decline when mild cognitive impairment was present, regardless of the variety of activities
(Hughes et al., 2013).
Howrey et al. (2015) used 18 years of the H-EPESE data (1993–2010) to describe
psychosocial, health, and lifestyle factors associated with maintaining adequate cognitive
function. They hypothesized that Mexican-origin older adults (65 or older) would have distinct
trajectories of cognitive function, which would vary depending on psychosocial and physical
health factors. Three measures on psychosocial factors were used: (a) depressed mood using the
Center for Epidemiologic Studies Depression Scale, (b) church attendance, and (c) social support
24

assessed with a two-item question measuring perceived social support. Cognition was assessed
using the MMSE as a continuous scale to capture covariates’ impact over time. Using latent class
models of three distinct cognitive trajectories—stable, slow decline, and rapid decline—the study
showed cognitive resilience was associated with church attendance, female gender, high school
education, and marriage (Howrey et al., 2015). This study used data from all respondents,
including those with cognitive impairment, which may have created bidirectional causality from
family or friends increasing social support due to cognitive decline. Nonetheless, it provides an
opportunity to replicate some analysis methods for understanding older adults without severe
cognitive decline as they transition from middle-old (75+) to old-old (85+) adulthood, as
described in Chapter 1.
Sharifian et al. (2019) selected data from the Washington Heights-Inwood Columbia
Aging Project (n = 548, 60–93 years old), a prospective longitudinal study of aging and dementia
in northern Manhattan (Sharifian et al., 2019). The study included non-Hispanic Whites (31%),
African Americans (41%), and Caribbean Hispanics of any race (28%) without dementia to
differentiate the racial and ethnic relationship between network characteristics and cognition.
Social network was characterized as the number of living children, relatives, and close friends.
Cognitive function composite scores were derived from four domains: episodic memory,
language, visuospatial skills, and speed and executive functioning. The neuropsychological tests
used for the composite score were (a) selective reminding test for episodic memory; (b) Benton
visual retention test, Rosen drawing test, and a subset from the Dementia Rating Scale for
visuospatial skills; (c) color trails test for speed and executive functioning; and (d) naming, letter
and category fluency, verbal abstract reasoning, repetition, and comprehension for language.
Using multiple regression analysis, Sharifian et al. (2019) found Caribbean Hispanics had a
25

greater proportion and number of family members in their networks when compared to the other
groups. Caribbean Hispanics had lower global cognitive scores, and no significant association
was found with social network characteristics. The authors pointed out that additional research
that is sensitive to cultural nuances such as living in ethnic enclaves, immigration, acculturation,
and social resources is necessary to better understand cognitive function in Caribbean Hispanics.
In reviewing the tables presented in this article, the Caribbean Hispanics had less educational
attainment compared to the other groups (Non-Hispanic White: 16 years; African American:
13.82 years; Caribbean Hispanic: 9.05 years). Education and race and ethnicity have been
previously linked to lower performance in neuropsychological tests (Acevedo et al., 2000;
Ardila, 1995; Rosselli & Ardila, 2003).
Rote et al. (2021) examined social support trajectories (emotional and instrumental) of
older Mexican Americans using H-EPESE data from 1993–94 and 2010–11 to understand the
changes in social support as dementia progresses. The study tested five hypotheses: (a) emotional
and instrumental support trajectories will be distinct; (b) living alone, immigrant status, and male
gender represent the greatest risk factors; (c) dementia trajectories will be well defined; (d)
women and immigrants have higher dementia risk in later life; and (e) low support creates an
increasing risk of dementia. The dependent variables were social support assessed by two
questions—having someone to talk to about deepest problems and being able to count on
someone in times of trouble—and likely dementia measured using the MMSE and instrumental
activities of daily living (IADL). “Likely dementia” meant that a respondent had a total MMSE
score between 0 and 23 with at least one IADL deficit. Those who had MMSE scores between 0
and 23 with no IADL deficits were identified as having “cognitive impairment with no IADL,”
whereas those with MMSE scores of 24 to 30 with no IADL deficits were considered to have “no
26

impairment.” Key independent variables included the number of people in the home and marital
status. Control variables were age, gender, education, nativity, and access to health care based on
having Medicaid. Using latent class models, trajectories of instrumental, emotional, and social
support needs and likely dementia were created to classify parallel changes of social support and
likely dementia during a 15-year period. Rote et al. (2021) found that those with likely dementia
had low social support and that living alone increased the likelihood of impairment and likely
dementia. The authors indicated the need to conduct research that takes a more comprehensive
approach to understanding the role of social engagement, including social support, to create
policies and programs to help combat and assist with dementia. Rote et al. (2021) reinforced the
importance of conducting research focused on persons with normal cognition to understand their
trajectory and avoid reverse causality.
Theoretical Frameworks Linking Social Engagement and Cognitive Impairment
This dissertation drew from two theoretical models used to understand the association
between social engagement and cognitive decline and the implications for age-friendly cities
policies. The first serves to provide a cause-and-effect context, albeit the research literature in the
area of social engagement shows considerable variation in constructs, as the articles noted here
demonstrate. The second framework offers a lens for understanding that although the focus of
this dissertation is late life, implications are relevant throughout life.
The NASEM (2020) consensus study report on social isolation and loneliness among
older adults developed a conceptual framework (hereafter called a “guiding framework”) to help
understand the various aspects of social connections in the structure of social determinants of
health. Although research in this area has not definitively confirmed the causal mechanisms, the
guiding framework conceptualizes the social connection pathways affecting health. The guiding
27

framework was developed to address issues of social isolation and loneliness, but it
acknowledges that these aspects of social relationships affect social engagement. The terms here
are key to understanding the guiding framework, shown in Figure 5 (NASEM, 2020):
 Loneliness: the perception of social isolation or the subjective feeling of being
lonely
 Mediators: also known as mechanisms or pathways; the factors that help explain
how social isolation or loneliness affects health outcomes
 Moderators: the factors that can influence the magnitude or direction of the
effect of social isolation or loneliness on health
 Social connection: an umbrella term that encompasses the structural, functional,
and quality aspects of how individuals connect to each other
 Social support: the actual or perceived availability of resources (e.g.,
informational, tangible, emotional) from others, typically in one’s social
network
28

Figure 5. Modified NASEM Guiding Framework
The guiding framework, as shown in Figure 5, is based on a hypothesis that assumes a
relationship exists among the individual, community, and society. In this framework, a thick line
indicates a hypothesized causal relationship and a thinner line suggests a potential relationship.
Social engagement is the outcome of social connections taking place at the individual level. We
can see that social connections have a direct impact on cognitive health (health impacts), but
cognitive decline can also affect social connections. Social connections affect risk factors (e.g.,
genetics) in a similar relationship with mediators (e.g., history of smoking). Mediators have a
potential bidirectional causal relationship with risk factors. Mediators have a causal relationship
with health impacts, and health impacts can potentially affect mediators. These individual-level
relationships take place in communities where things such as pollution or ethnic enclaves are
influenced by the larger society (health care, family structures, policies). This guiding framework
makes it clear that capturing the effects of social engagement is complex and dynamic, with
29

public policies having a direct influence on the ability of individuals to engage in and experience
cognitive stimulation.
Social engagement is lifelong and dynamic. The social trajectory model provides a
framework to understand the link between early childhood experiences and adult health
outcomes, including late-life outcomes. The social trajectory model as shown in Figure 6
proposes that adult health outcomes are not directly determined by early life social conditions.
Adult social conditions present an opportunity to reshape childhood socioeconomic disadvantage
to improve adult health outcomes (Berkman et al., 2011).
Figure 6. Social Trajectory Model
Frameworks to Support an Aging Society
The projected demographic shift along with the expected public health crisis of dementia
have led to two interrelated global movements: age-friendly cities and dementia-friendly
communities. These global movements provide an opportunity to create environments at the
community and societal levels that consider culture in the development of policies and programs.
As such, the relevance of the findings of this study will be considered through the lens of these
interdependent global movements that provide opportunities for innovation.
Almost two decades ago, the WHO adopted a policy framework for communities to start
creating action plans that promoted healthy and active aging. The concept of active aging
consists of physical, mental, and social well-being and has been defined as “the process of
30

optimizing opportunities for health, participation, and security in order to enhance equality of life
as people age” (WHO, 2002). Emerging from the 2002 WHO active aging framework was the
creation in 2007 of the age-friendly cities framework. The core goal of the framework is to
encourage the establishment of policies, structures, and services that are accessible and inclusive
of older adults who have various needs and capacities. It highlights domains that represent the
cornerstone features necessary to support active aging: (a) housing, (b) social participation, (c)
respect and social inclusion, (d) civic participation and employment, (e) communication and
information, (f) community support and health services, (g) outdoor spaces and buildings, and
(h) transportation. The age-friendly cities framework was built on a life course perspective to
include people of all ages in the process and acknowledge that experiences throughout life shape
individuals, creating increased diversity in late life (WHO, 2007).
Scharlach and Lehning (2013) described the age-friendly characteristics necessary to
foster lifelong activities, support basic needs, maintain significant relationships, and promote
socially meaningful engagement in personal and social activities with opportunities to develop
new interests and sources of satisfaction. The authors pointed out that U.S. public policy has not
given much attention to the community context in which older adults reside. The lack of
consideration of the social environment tends to foster social isolation and exacerbate social
exclusion. Furthermore, the ability to remove barriers to social relationships not just for older
adults but also for individuals of all ages through formal organization participation or informal
interactions creates a mechanism for social engagement. Scharlach and Lehning (2013)
emphasized that the empirical literature indicates social inclusion (social integration, social
support, and access to resources) can only be supported “through physical infrastructure
31

improvements such as walkable neighborhoods, mobility options, and adequate housing for
persons with diverse needs and abilities.”
Dementia Friendly America was launched after the 2015 White House Conference on
Aging to support people with dementia to continue as active and thriving members of their
community for as long as possible (World Dementia Council, 2020). In the United States, the
dementia-friendly initiative emphasizes social aspects instead of focusing on medical issues
(Turner & Morken, 2016). As shown in Table 4, Turner and Morken (2016) compared the
domains of age-friendly and dementia-friendly communities. They found overlaps in the areas
that promote independence in late life through supportive environments. Age-friendly
communities may not always incorporate dementia-friendly approaches. But there is a clear need
to have a process that involves community members, key stakeholders, and political commitment
with short- and long-term goals to create age-friendly communities that incorporate dementia-
friendly domains into actionable plans. This dissertation was built on the premise that age-
friendly communities through the domain of social participation have a platform to support and
enact policies and programs to position social engagement as a key and necessary feature with
direct positive benefits for older adults and societies.
Table 4. Cross-Referencing Age-Friendly Domains and Dementia-Friendly Sectors
WHO Age-Friendly Domains Dementia-Friendly Sectors
Outdoor Spaces and Buildings ꞏ Transportation, Housing, and Public Spaces
ꞏ Emergency Planning and First Response
Transportation ꞏ Transportation, Housing, and Public Spaces
ꞏ Emergency Planning and First Response
Housing ꞏ Transportation, Housing, and Public Spaces
ꞏ Emergency Planning and First Response
ꞏ Independent Living
ꞏ Memory Loss Supports and Services
Social Participation ꞏ Legal and Advance Planning Services
32

ꞏ Banks and Financial Services
ꞏ Businesses
ꞏ Neighbors and Community Members
ꞏ Communities of Faith
ꞏ Independent Living
Respect and Social Inclusion ꞏ Legal and Advance Planning Services
ꞏ Banks and Financial Services
ꞏ Businesses
ꞏ Neighbors and Community Members
ꞏ Communities of Faith
ꞏ Independent Living
Civic Participation and Employment ꞏ Businesses
ꞏ Communities of Faith
ꞏ Independent Living
Communication and Information Woven into each sector as dementia-friendly
communication strategies
Community and Health Services ꞏ Care Throughout the Continuum
ꞏ Memory Loss Supports and Services
ꞏ Emergency Planning and First Response

33

CHAPTER 3: METHODOLOGY
This chapter describes the methods and research design of this dissertation study. It starts
with the research question, followed by an overview of the longitudinal cohort study that
provided the dissertation’s data. Specifically, the sample and population, instrumentation, data
collection, and data analysis were used to examine and inform public policies and practices to
create age-friendly cities. The research question guiding this study was: What features of social
engagement in late life are associated with lower prevalence of likely dementia? The hypothesis
was that late-life social engagement serves as a protective cognitive factor.
Sample and Population
In 1992, the National Institute on Aging first funded the Hispanic Established Population
for the Epidemiological Study of the Elderly, commonly known as H-EPESE or Hispanic-
EPESE, and has renewed funding through 2021. H-EPESE is a longitudinal cohort study of
community-dwelling older Mexican Americans in Arizona, California, Colorado, New Mexico,
and Texas (Office of Extramural Research, 2020). These five states were selected based on 1990
U.S. census data that indicated that 85% of their population self-identified as Mexican American
and of Mexican origin (Markides, 1999). Although the H-EPESE was modeled after existing
EPESE studies in New Haven, East Boston, Iowa, and North Carolina, it generally drew from the
North Carolina-Duke EPESE. The H-EPESE sought to extend the information available on
Mexican American older adults’ health and socioeconomic characteristics.
Markides (1999) provided a detailed description of the multistage area probability
sampling approach used to identify Mexican Americans. Counties were selected based on the
number of Mexican Americans older adults living there. The same process was used to select
300 census tracts that became the primary sampling units. The block groups of each census tract
34

were used to conduct a systematic random selection process (multiplying a random integer by the
number of households) for the enumeration process. Each sampling unit had 400 households in a
block group, and when necessary, block groups were aggregated to reach minimum sampling
targets. Interviewers were sent door to door to count older adults residing in either single, family,
or apartment buildings (no commercial units) to create enumeration lists with 175 households. A
household with five or more older adults was marked as a group home and excluded from the
study. Sample weights were applied at the census tract level and poststratified to fit each state’s
distribution of Mexican Americans aged 65 or older. In 2004–05, a probability sample of 902
Mexican American older adults of similar age, 75 or older, was added to increase the study
sample size to 2,069 (Cantu & Markides, 2019). Data from the H-EPESE can be accessed by the
general public and downloaded from the National Archive of Computerized Data on Aging.
Table 5 (adapted from Cantu & Markides, 2019) indicates the sample sizes collected during each
study wave.
Table 5. H-EPESE Sample Size by Data Collection Year
Wave Year Respondents Note
1 1993–94 3,050
2 1995–96 2,438
3 1998–99 1,980
4 2000–01 1,685
5 2004–05 2,069 Dissertation baseline
6 2006–07 1,542 Follow-up period
7 2010–11 1,078
8 2012–13 744
35

9 2016 480
Instrumentation
To this day, the H-EPESE uses personal interviews, questionnaires, and physical
assessments to collect data on physical health, mental health, and functional impairment
(Markides, 1999).
Selected Variables
Cognition, the primary outcome measure, was assessed using the Folstein MMSE.
Researchers have found the MMSE’s performance is influenced by education, age, and ethnicity
(Tombaugh & McIntyre, 1992). The MMSE is the most commonly used cognitive field tool to
screen for cognitive impairment by assessing the areas of orientation, registration, recall,
calculation, attention, naming, repetition, comprehension, reading, writing, and drawing
(Mitchell, 2017). Scores are used to determine the current status of cognitive impairment. Older
adults with low education levels tend to have lower scores when compared to non-Latino Whites.
Researchers have addressed this by using lower cutoff scores, controlling for age and education
to improve specificity and sensitivity (Black et al., 1999; Crum et al., 1993; Matallana et al.,
2011; Parker & Philp, 2004; Whitfield et al., 2000). When using longitudinal data, such as the H-
EPESE, literacy tends to be a stable variable, and declines in MMSE scores over time indicate a
true cognitive decline likely related to dementia (Markides et al., 2009).
The MMSE’s reliability is stronger when individuals have moderate to severe impairment
and is considered to have “modest qualities in screening for mild cognitive disturbances” (Bour
et al., 2010). The MMSE is not diagnostic tool but can rule out possible or probable cognitive
impairment (Mitchell, 2017). Persons with dementia experience a decline in their score of about
3 to 4 points per year after memory problems (C. M. Clark et al., 1999; Cockrell & Folstein,
36

2002). Several research studies have found the MMSE to accurately predict a diagnosis of
dementia for individuals showing poor cognitive performance (Bour et al., 2010; Tierney et al.,
2000). Studies have indicated that MMSE findings of cognitive impairment correlate with
diagnostic brain imaging tests such as single-photon emission computed tomography (DeKosky
et al., 1990) and magnetic resonance imaging (Bondareff et al., 1990; Garcia-Diaz et al., 2014).
The independent variables for social engagement were identified based on the GBHC
social engagement framework description of social engagement components, as previously
described in Table 3. The selected variables, including the survey questions, are listed in Table 6.
Table 6. List of Social Engagement Variables
Domain Variable
Name
Variable Question
SE Structural Components
Frequency of contact SEEMON5

How many of your (#CHILDREN IN Q.D1)
children do you see at least once a month?
RSEEMON5 How many of these (# RELATIVES IN
Q.D5) relatives do you see at least once a
month?
FSEEMON5 How many close friends do you see at least
once a month?
Church attendance EE52 How often do you go to church or religious
services?
Presence or absence: family
or friends, partner, spouses,
neighbors
MARSTAT5 Are you presently married, divorced,
separated, widowed, or never married?
(include common law marriage under
married)
F4NA5 How many relatives live in neighborhood but
not with you?
F4NB5 How many friends live in your
neighborhood?
LIVALON5 Person lives alone
SE Functional Components
Emotional support COUNTON5 Can you talk about your deepest problems
with at least some of your family or friends
most of the time, some of the time, or hardly
ever?
37

Health behaviors have been identified as possible mediators in the causal pathway of the
development of diseases that cause cognitive impairment, with possible bidirectional relationship
with social support (NASEM, 2020). The health behaviors listed in Table 7 were selected based
on their inclusion of lifestyle factors that have been identified as risk factors for cognitive
impairment (Livingston et al., 2020).
Table 7. List of Health Behavior Variables
Health Behaviors
Health care
utilization
VISITMD5 Created variable (0 = no, 1 = yes) of at least one visit
to a medical doctor including physician assistant or
nurse practitioner.
Cigarette use YSMOKE53 Do you smoke cigarettes now?
Alcohol use ZALC52 In the past month, have you had any beer, wine, or
liquor?
The chronic health conditions listed in Table 8 were collected via self-reported measures.
These variables were selected based on existing literature that indicates cardiovascular
conditions and depression are risk factors for cognitive impairment (Livingston et al., 2020;
WHO, 2012).
TALK5 In times of trouble, can you count on at least
some of your family or friends most of the
time, some of the time, or hardly ever?
CLOSREL5 How many relatives do you have that you
feel close to—that you feel at ease with, can
talk to about private matters, or can call?
(INCLUDE siblings, in-laws, EXCLUDE
spouse and children.)
CLOSFRN5 Other than members of your family, how
many close friends do you have—people that
you feel at ease with, can talk to about
private matters, or can call on for help?
SE Quality Components
Sense of social well-being CC53 Now please think about your life as a whole.
How satisfied are you with it?
38

Table 8. List of Health Condition Variables
Depression status was determined using Radloff’s (1977) Center for Epidemiologic
Studies Depression Scale, a self-report measure that screens for the presence of depressive
symptoms. The scale consists of 20 questions about how often someone has experienced
depressed mood, guilt, worthlessness, helplessness, hopelessness, psychomotor retardation, lack
of appetite, and sleep problems during the past week. Respondents select from a 4-point Likert
scale, with potential total scores ranging from 0 to 60. A score of 16 or more is considered
indicative of clinical depression (Radloff, 1977). Sociodemographic variables were derived from
information on age based on the respondent’s date of birth, gender as either female or male as
recorded by the screener, and highest grade of school completed.
Data Analysis
All data analyses were conducted using the statistical software Stata 14 (StataCorp,
2015). Anonymized data were obtained from the H-EPESE research team. The dataset was
delivered as single files for each data collection wave. To prepare the database for cross-
sectional analysis, Wave 5 (baseline) was merged with Wave 6 (follow-up) and their
Chronic Health Conditions
Diabetes MDIAB51 Have you ever been told by a doctor that you have
diabetes, sugar in your urine, or high blood sugar?
Hypertension KHYPER51 Has a doctor ever told you that you have high blood
pressure?
Cardiovascular ICARDI51 Have you ever been told by a doctor or other health
care professional that you had any of the following
conditions: had or suspected a heart attack, coronary,
myocardial infarction, or coronary thrombosis?
Stroke JSTROK51 Have you ever been told by a doctor or other health
care professional that you had any of the following
conditions: had or suspected a stroke, blood clot in
the brain, or brain hemorrhage?
Feelings and
depression
CASE5 Sum scored of the Center for Epidemiologic Studies
Depression Scale
39

corresponding mortality data. Participants without survey data during baseline were removed
from the database (n = 2,069). The dataset has three types of respondents: respondent only, proxy
only, and long proxy that includes proxy and respondent. All participants whose baseline data
was collected via proxy and long proxy were removed from the database. Proxy responses do not
include participants’ MMSE information (dependent variable) and in the case of long proxy,
surveys may include responses from informants. To address possible reverse causality between
social engagement and cognition, participants with a MMSE score of less than 21 were excluded.
This cutoff score was used based on previous H-EPESE research that identified participants as
having normal cognition using a single cutoff score of 21 (Downer, Al Snih, et al., 2019; Garcia
et al., 2018; Raji et al., 2004). Also excluded were participants who during follow-up were noted
as deceased, proxy, lost to follow-up, or missing total MMSE score.
To compare the association of cognition and social engagement, a dummy variable for
cognition was created using the follow-up MMSE total scores as a reference. Respondents with a
score of 21–30 were coded as normal cognition (0) and those with a score of 0–20 as likely
dementia (1). Finally, all follow-up data and other variables not selected for analysis were
excluded. As indicated in the analytic sample Figure 7, the final sample was 863.
40

Figure 7. Flowchart of Sample Selection

Binary variables were created for all selected variables to allow for comparison of
characteristics among older adults. Marital status was recoded as 0 (not married) and 1
(married). The four binary variables related to perceived emotional support—most of the time
can you count on someone, most of the time can you talk to someone, number of relatives
available to talk about private matters, and number of friends available to talk about private
matters—were summed to create a binary variable of having emotional support from all four
sources. The binary variables for health behaviors were summed to indicate the presence of one
or more risky health behaviors, and chronic health conditions including depression were summed
41

to indicate the presence of two or more chronic health conditions. A correlational analysis was
performed using a generous p-value threshold of less than .10 to have a general understanding of
strength of the relationship of social engagement and cognition.
The association of the dependent variable to the independent variable was calculated
using binary logistic regressions with odds ratio. Logistic regressions assess the probability that
the independent variables predict associations with the dependent variable categories, in this case
normal cognition or likely dementia. The odds ratio “compare the relative odds of the occurrence
of the outcome,” likely dementia, given exposure to the social engagement and other
independent variables (Szumilas, 2010). An odds ratio greater than 1 indicates a positive
relationship (higher odds) meaning that exposure to the independent variables is associated with
higher odds of likely dementia. If the odds ratio is less than 1, there is a negative relationship
(lower odds) indicating that exposure to the independent variables lowers the odds of likely
dementia (Szumilas, 2010). All analysis results are presented in Chapter 4 and a printout of the
Stata do file can be found in Appendix B.
Institutional Review Board
This study was approved on July 2, 2020, by the University of Southern California
Institutional Review Board (approval number UP-20-00457) as an exempt study under the
project title “Social Participation and Cognition: Age-Friendly.” See Appendix A for copy of the
Institutional Review Board approval letter.

42

CHAPTER 4: DATA REPORTING AND ANALYSIS
This chapter presents the data analysis results related to answering the research question:
What features of social engagement in late life are associated with lower prevalence of likely
dementia? It includes the results of the descriptive and correlation analysis, along with the
logistic odds ratio regressions. The do files used to create the database and variables are included
in Appendix B.
Descriptive Statistics
Nearly 18% of the analytic sample was characterized as having likely dementia based on
the follow-up assessment completed 6–7 years postbaseline, when the mean age was 80. A larger
proportion of participants were women (63.6%), and the full sample’s mean education level was
at the sixth-grade level. Compared to women (15%), more men (22%) were categorized as
having likely dementia. The 85 or older category had more women (n = 97), with about 26%
found to have likely dementia at follow-up, along with about 32% of men in the same age cohort.
Participants with 0 to 4 years of formal education were slightly more prevalent (n = 348) than the
group with 9 or more years of education (n = 228). In the latter group, women were found to be
5% more likely to have likely dementia: 28% of women vs. 23% of men. Table 9 shows the
frequency and percentage (frequency and gender total) of the dummy variables.
Table 9. Descriptive Statistics
Dummy Variables
Likely Dementia (n = 153) Normal Cognition (n = 710)

Male
(n = 69)
Female
(n = 84)

Male
(n = 245)
Female
(n = 465)
n n % n % n n % n % n
Sociodemographics

Age 75–79 427 27 39.13 30 35.71 57 132 53.88 238 51.18 370
Age 80–84 279 23 33.33 29 34.52 52 72 29.39 155 33.33 227
Age 85–109 157 19 27.54 25 29.76 44 41 16.73 72 15.48 113
Education 0–4 years 348 42 60.87 46 54.76 88 102 41.63 158 33.98 260
Education 5–8 years 287 19 27.54 20 23.81 39 78 31.84 170 36.56 248
Education 9+ years 228 8 11.59 18 21.43 26 65 26.53 137 29.46 202
Social Engagement: Structural Components

43

No children seen monthly 103 5 7.35 9 10.71 14 22 8.98 67 14.41 89
1–2 children seen monthly 327 20 29.41 30 35.71 50 92 37.55 185 39.78 277
3+ children seen monthly 432 43 63.24 45 53.57 88 131 53.47 213 45.81 344
No close family seen monthly 314 35 51.47 35 41.67 70 98 40.00 148 32.31 246
1–2 close family seen monthly 356 24 35.29 30 35.71 54 92 37.55 211 46.07 303
3+ close family seen monthly 182 9 13.24 19 22.62 28 55 22.45 99 21.62 154
No close friends seen monthly 275 25 36.76 28 33.73 53 85 35.42 137 30.04 222
1–2 close friends seen monthly 339 22 32.35 37 44.58 59 83 34.58 197 43.20 280
3+ close friends seen monthly 233 21 30.88 18 21.69 39 72 30.00 122 26.75 194
No regular church attendance 144 18 26.09 19 22.62 37 46 18.78 61 13.15 107
Regular church attendance 274 26 37.68 25 29.76 51 83 33.88 140 30.17 223
Almost weekly or weekly church attendance 444 25 36.23 40 47.62 65 116 47.35 263 56.68 379
Married 394 48 69.57 18 21.43 66 173 70.61 155 33.48 328
Has family or friends as neighbors 733 60 86.96 71 84.52 131 208 84.90 394 84.73 602
Alone in household 260 13 18.84 33 39.29 46 41 16.73 173 37.20 214
1 other in household 381 33 47.83 32 38.10 65 121 49.39 195 41.94 316
2+ in household 222 23 33.33 19 22.62 42 83 33.88 97 20.86 180
Social Engagement: Functional Components

Most times has someone to count on 688 51 73.91 73 86.90 124 187 76.64 377 81.43 564
Most times has someone to talk to 608 44 64.71 66 78.57 110 166 68.03 332 71.55 498
1+ family feel close 643 40 58.82 58 69.05 98 186 75.92 359 77.87 545
1+ friends feel close 602 46 68.66 59 71.08 105 160 66.67 337 73.58 497
Has someone to count on, talk to, feels close
to family and friends
381 25 36.23 39 46.43 64 95 39.58 222 47.74 317
Social Engagement: Quality Components
Completely satisfied with life 398 30 44.12 36 42.86 66 123 50.20 209 44.95 332
Very satisfied with life 354 28 41.18 35 41.67 63 93 37.96 198 42.58 291
Somewhat or not satisfied with life 110 10 14.71 13 15.48 23 29 11.84 58 12.47 87
Current Health Behaviors

Lacks regular health care 95 12 17.39 11 13.41 23 30 12.24 42 9.05 72
Current smoker 53 8 11.59 4 4.82 12 24 9.88 17 3.74 41
Drinks alcohol 150 17 24.64 5 5.95 22 71 29.10 57 12.26 128
1+ risk health behavior 265 30 43.48 20 23.81 50 103 42.04 112 24.09 215
Health Conditions

Diabetes 268 20 29.41 32 38.10 52 71 28.98 145 31.25 216
Hypertension 546 36 53.73 57 67.86 93 146 59.59 307 66.16 453
Heart disease 203 24 34.78 16 19.05 40 55 22.54 108 23.43 163
Had or suspected stroke 51 7 10.29 6 7.14 13 14 5.71 24 5.17 38
Had or suspected cardiovascular disease 53 7 10.29 5 6.02 12 18 7.35 23 4.96 41
Depression 94 8 11.59 14 16.67 22 11 4.49 61 13.12 72
2+ chronic conditions 372 38 55.07 44 52.38 82 87 35.51 203 43.66 290
Structural components of social engagement were more prevalent in the normal cognition
group. Differences of more than 5% between men and women occurred in the likely dementia
group for the number of children seen monthly, no family members seen monthly, close friends
seen monthly, church attendance, married status, and living with one other person in the
44

household. In the normal cognition group, differences were found in the dummy variables of one
to two friends seen monthly, one to two family members seen monthly, no children seen
monthly, married, living alone, and living with two or more other people in the household.
Overall relative to men, women reported being more engaged with friends, more likely to be
unmarried, and more likely to live alone.
The functional components in the likely dementia group did not differ more than 5%. In
the normal cognition group, the dummy variables for perceived support of having some to count
on, someone one to talk to, and their sum total of having a confidant all had higher percentages
for women. Also, women were more likely than men to report having one or two family
members who feel close and one or two friends who feel close. Overall, more women reported
having at least one person in each functional component variable.
The quality component variable of life satisfaction indicated that more women (more than
5% difference) in the normal cognition group reported being very satisfied with life. Otherwise,
no notable differences were found in the normal cognition and likely dementia groups. As far as
current health behaviors, more men reported regularly drinking alcohol in both groups. In
general, men in both groups were more likely to engage in one or more risk health behavior.
The health conditions among those with likely dementia were more common in men, with
a difference of more than 5% found in heart disease. The normal cognition group had higher
percentages of women reporting hypertension, depression, and two or more chronic conditions.
Correlation Analysis
Table 10 indicates the correlational analysis performed using the categorial and nominal
variables in the database. The independent variables had a positive strong correlation to
cognition, meaning as one increased, the other also increased. These variables were education
45

level, church attendance, being able to count on someone, and life satisfaction. Other variables
found to be highly suggestive of a positive correlation were having someone to talk to, number
of close family members, alcohol, and cardiovascular disease.
Depression was found to have an inverse or negative correlation, meaning that as one
variable increased, the other decreased. Other variables found to have an inverse suggestive
correlation were having someone to talk to, smoking, and regular doctor visits. Regular doctor
visits were found to have a negative relationship in men but a positive relationship in women.
Due to the high correlation with education level and age, the sociodemographic variables were
introduced into logistic regression models last to mitigate the effects of their strong correlation
with the other variables.
Table 10. Correlations between Cognition and Covariates
Variables Both Genders
(n = 863)
Men (n=314) Women (n‐549)
r p N M r p N M r p N M
Sociodemographics



Age ‐.039 .257 863 80.6 ‐.026 .647 314 80.5 ‐.047 .275 549 80.6
Education Level .379 .000 863 6.03 .310 .000 314 5.77 .416 .000 549 6.18
Social Engagement: Structural
Kids seen monthly ‐.033 .334 802 3.41 .082 .146 305 3.57 ‐.091 .033 497 3.32
Close family seen monthly .044 .197 855 1.65 .078 .167 313 1.64 .022 .616 542 1.65
Close friends seen monthly .040 .248 847 2.14 .006 .923 308 2.29 .073 .088 539 2.06
Church attendance .082 .016 863 3.17 ‐.018 .755 314 2.97 .128 .002 549 3.29
Marital status ‐.016 .631 863 2.59 ‐.014 .804 314 1.82 ‐.056 .190 549 3.02
Family as neighbors ‐.002 .956 862 1.51 .055 .332 314 1.50 ‐.037 .381 549 1.51
Friends as neighbors .015 .664 859 2.08 .068 .233 314 2.11 ‐.013 .758 549 2.07
Number in household ‐.017 .616 863 1.96 ‐.051 .370 314 2.17 .020 .645 549 1.84
Social Engagement: Functional
Can count on someone ‐.098 .003 863 1.27 ‐.079 .160 314 1.33 ‐.105 .014 549 1.24
Can talk to someone ‐.064 .060 863 1.40 ‐.096 .089 314 1.47 ‐.035 .418 549 1.36
Close family .055 .105 858 2.27 .101 .073 313 2.22 .028 .509 545 2.29
Close friends .035 .308 848 2.49 .014 .813 307 2.67 .060 .167 541 2.39
Social Engagement: Quality
Satisfied with life ‐.104 .002 863 1.68 ‐.055 .329 314 1.67 ‐.138 .001 549 1.69
Current Health Behaviors



Regular doctor visits .038 .266 860 0.89 ‐.085 .134 314 0.87 .113 .008 546 0.90
Smoking ‐.061 .074 850 0.06 ‐.048 .393 312 0.10 ‐.062 .150 538 0.04
Alcohol .107 .001 862 0.17 .165 .003 313 0.28 .091 .033 549 0.11
Health Conditions



Diabetes .000 .995 863 1.70 ‐.068 .232 314 1.73 .045 .296 549 1.69
Hypertension ‐.021 .536 863 1.74 ‐.060 .285 314 1.86 .013 .761 549 1.67
46

Heart failure or disease .013 .704 863 1.80 .051 .366 314 1.77 ‐.008 .847 549 1.82
Cardiovascular .032 .348 860 1.94 .133 .018 313 1.92 ‐.058 .177 547 1.95
Stroke .016 .623 861 1.94 .051 .369 313 1.93 .014 .751 548 1.95
Depression ‐.111 .001 863 7.46 ‐.062 .273 314 5.89 ‐.150 .000 549 8.36
Logistic Regression: Odds Ratios
Logistic regression odds ratios (OR) were calculated for men, women, and the full
sample to compare and contrast the relationship of variables. The complete logistic regression
tables are presented in Appendix C. The OR findings are summarized in Table 11. Most of the
social engagement variables had null findings, but a few distinctive associations emerged for
men and women.
In the male sample, one social engagement variable had a statistically significant
association even after adding health behaviors and health conditions, but it did not sustain
significance after sociodemographics variables were added to the OR model. Frequency of
contact indicated by the number of family members (not including children) seen monthly was
associated with reduced likely dementia by 63% (OR = 0.361; Model 3). The presence of two or
more chronic conditions (OR = 2.579; Model 4) increased the odds of likely dementia. Men with
an education level of ninth grade or higher (OR = 0.281) had 73% lower risk of likely dementia.
The female sample analysis revealed that two social engagement variables, frequency of
contact with family and being married, remained statistically significant after adding the
covariates. Similar to men, frequency of contact with family lowered the risk of likely dementia,
by 49% (OR = 0.511; Model 4). Women respondents who reported being married had 53% lower
odds of likely dementia (OR = 0.474; Model 4). Having a fifth-grade (OR = 0.365) or higher (OR
= 0.526) education reduced the odds of likely dementia.
Table 11. Odds Ratio Summary: Men, Women, and Full Sample
Model 1: Social Engagement
OR p 95% CI
47

Men (n = 314)
3+ close family seen monthly 0.361 .029 0.14, 0.90
Women (n = 549)
1–2 close family seen monthly 0.539 .050 0.29, 1.00
Married 0.434 .013 0.22, 0.84
Full Sample (n = 863)
1–2 close family seen monthly 0.610 .029 0.39, 0.95
3+ close family seen monthly 0.581 .052 0.34, 1.01
Model 2: Social Engagement and Health Behaviors
Men
3+ close family seen monthly 0.355 .027 0.14, 0.89
Women
1–2 close family seen monthly 0.539 .050 0.29, 1.00
Married 0.433 .013 0.22, 0.84
Full Sample
1–2 close family seen monthly 0.612 .030 0.39, 0.95
3+ close family seen monthly 0.574 .048 0.33, 0.99
Model 3: Social Engagement, Health Behaviors, and Health
Conditions
Men
3+ close family seen monthly 0.361 .030 0.14, 0.90
2+ chronic conditions 2.134 .013 1.17, 3.87
Women
1–2 close family seen monthly 0.521 .040 0.28, 0.97
Married 0.438 .015 0.23, 0.85
Full Sample
1–2 close family seen monthly 0.590 .021 0.38, 0.92
3+ close family seen monthly 0.563 .041 0.32, 0.98
2+ chronic conditions 1.581 .016 1.09, 2.30
Model 4: Social Engagement, Health Behaviors, Health
Conditions, and Sociodemographics
Men
3+ close family seen monthly 0.413 .071 0.16, 1.08
2+ chronic conditions 2.579 .003 1.37, 4.85
Education 9+ grade level 0.281 .004 0.12, 0.67
Women
1–2 close family seen monthly 0.511 .039 0.27, 0.97
Married 0.474 .032 0.24, 0.94
Education 5th–8th grade level 0.365 .001 0.20, 0.68
Education 9+ grade level 0.526 .053 0.27, 1.01
Full Sample
1–2 close family seen monthly 0.633 .052 0.40, 1.00
48

3+ close family seen monthly 0.603 .080 0.34, 1.06
2+ chronic conditions 1.667 .009 1.13, 2.45
Age 75–79 0.608 .024 0.39, 0.94
Age 85+ 1.658 .047 1.01, 2.73
Education 5th–8th grade level 0.443 .000 0.29, 0.69
Education 9+ grade level 0.451 .002 0.27, 0.74
In the analysis with the full analytic, sample the frequency of contact with family (OR =
0.052; Model 4) was found to be statistically significant in all regression models, decreasing the
odds of likely dementia. Reporting two or more chronic health conditions increased the odds of
likely dementia in all regression models (OR = 1.667; Model 4). Age reduced the odds of likely
dementia for participants aged 75–79 (OR = 0.608) and increased the odds for those aged 85–95
(OR = 1.658). Having a fifth-grade education (OR = 0.443) or higher (OR = 0.451) lowered the
odds of likely dementia.

49

CHAPTER 5: DISCUSSION AND POLICY RECOMMENDATIONS
In the United States, more than 6 million people live with dementia, and by 2050, the
number of people with dementia is projected to reach nearly 13 million (Alzheimer’s
Association, 2021). Unfortunately, the growth rate is disproportionate, and not all racial and
ethnic groups are affected equally. Latinos are affected by dementia at a rate 1.5 times higher
than non-Latino Whites (Alzheimer’s Association, 2021). Yet few research studies have focused
on Latinos, particularly Mexican Americans, the largest Latino group in the United States
(Gonzalez, 2020).
A comprehensive Lancet Commission systematic review of the dementia research
literature identified 12 risk factors that include the lack of social contact as affecting the
development of dementia (Livingston et al., 2020). Social engagement was described in their
first review of the literature as “a necessary condition for wellbeing throughout life” (Livingston
et al., 2017). Their report supports the understanding that the conditions in which people are
born, grow, work, live, and age and the broader set of forces and systems often referred to as the
social determinants of health shape the conditions of daily life, including social relations
(Dahlgren & Whitehead, 1991; WHO, 2019).
Given the current understanding of dementia and the dearth of Latino dementia research,
this dissertation examined social engagement during late life among older Mexican Americans
without severe cognitive impairment. Specifically, to answer the research question of what
features of social engagement in late life are associated with likely dementia, data from the
longitudinal H-EPESE were analyzed. The H-EPESE allowed analysis to be conducted using
data collected in 2004–05 (Wave 5), when participants were 75 or older.
50

Social engagement has been described as having multidimensional features inclusive of
structural, emotional, and quality components (Carlson et al., 2017). Based on the literature
review investigating social engagement and cognition in older Latinos in the United States, this
is the first study to take a comprehensive approach to capturing the various features of social
engagement. Each feature of each component was used to identify social engagement elements in
the H-EPESE data. Other studies using the H-EPESE data mostly have relied on two functional
component variables (emotional support) to describe social engagement and one structural
component (church attendance) for participation. The review of the baseline variables revealed
that eight variables represented structural social engagement, four for functional, and one for
quality.
In recognition of the social determinants of health and the role of social engagement, this
dissertation expanded the understanding of social engagement with the NASEM (2020) social
isolation and loneliness conceptual framework. This guiding framework specified additional
connections that create direct and indirect pathways that influence social engagement and
connections through risk factors, mediators, and health. These connections led to identifying
covariates related to health behaviors, chronic health conditions, and sociodemographic factors.
This dissertation’s primary outcome, cognitive impairment, was assessed using the
Folstein MMSE (Folstein et al., 1975). Although older adults with lower educational attainment
tend to have lower MMSE scores when compared to non-Latino Whites, field practitioners
heavily utilize the MMSE for its ability to assess orientation, registration, recall, calculation,
attention, naming, repetition, comprehension, reading, writing, and drawing (Crum et al., 1993).
Although researchers generally acknowledge the potential biases of the MMSE toward lower
total scores, they have addressed the lack of sensitivity to education level through the use of a
51

lower cutoff score that improves specificity and sensitivity (Black et al., 1999; Crum et al., 1993;
Matallana et al., 2011; Parker & Philp, 2004; Whitfield et al., 2000). Consistent with prior H-
EPESE research studies that utilized the MMSE variable (Downer, Al Snih, et al., 2019; Garcia
et al., 2018; Raji et al., 2004), the selection of participants and classification of cognitive status
in this study relied on a likely dementia cutoff score of less than 21.
The original baseline sample of 2,069 was reduced to 863 after deleting proxy interviews
and participants who had likely dementia at baseline to minimize reverse causality. Those who
were deceased, lost, or declined to participate at the 6-year follow-up were removed from the
database. The follow-up MMSE scores were used to classify participants as having likely
dementia (n = 153) or normal cognition (n = 710). The association of likely dementia to social
engagement was calculated using four models of binary logistic regressions. An odds ratio
greater than 1 indicated a positive relationship (higher odds), meaning that exposure to social
engagement was associated with higher odds of likely dementia. If the odds ratio was less than 1,
there was a negative relationship (lower odds) indicating that exposure to social engagement
lowered the odds of likely dementia (Szumilas, 2010).
Traditional characterizations of Mexican Americans tend to depict the culture as
patriarchal, with men and women in traditional gender roles (Keefe et al., 1979; Landale et al.,
2006; Markides et al., 1986; W. A. Vega, 1990). Therefore, the statistical analysis was conducted
on men, women, and the full sample individually to ascertain gender-specific nuances in social
engagement. Although most associations with the social engagement variables were null, some
gender-specific differences were noted.
The perceived support of feeling close to three or more family members was a positive
factor found to be statistically significant (p = .05) in the logistic regression analyses for all
52

subsamples, ranging from lower likely dementia risk of 63% for men to 46% for women and
43% for the full sample. Research on Mexican Americans has underscored the importance of
kinship ties, referred to in the literature as familismo, providing socioemotional support that
often leads to beneficial health outcomes (Grebler et al., 1970; W. A. Vega, 1990). R. J. Angel
and Angel (2009) found that in Latino culture, family is socially protective for older adults. A
protective feature based on these findings seems to extend to cognitive decline.
Married women had a 53% lower likelihood of likely dementia after all covariates were
added to the model. Sommerlad et al.’s (2018) systematic review and meta-analysis of studies on
the association of marital status and dementia found that being married lowered the risk of
developing dementia. Studies using H-EPESE data have also found that being married affords
women a protective effect or slower cognitive decline trajectory (Black et al., 1999; Monserud,
2019).
Similar to the general literature on dementia and other Latino-specific studies, including
Mexican Americans (Black et al., 1999; Downer, Garcia, et al., 2019; Howrey et al., 2015; Rote
et al., 2021), education emerged as associated with lower odds of likely dementia, whereas
having two or more chronic conditions was linked with higher odds of likely dementia.
The absence of frequency of contact can be characterized as social isolation. Social
isolation has been found to be increasing in the growing older adult population (Holt-Lunstad,
2020; Klinenberg, 2016; NASEM, 2020; Valtorta et al., 2016). Social isolation is defined as “the
objective state of having few social relationships or infrequent social contact with others” (Holt-
Lunstad et al., 2017; NASEM, 2020). Almost 1 in 4 community-dwelling Americans aged 65 or
older are socially isolated (NASEM, 2020). The cognitive stimulation of engagement with a
spouse, family, friends, or coworkers seems to be beneficial for general well-being and cognitive
53

function (National Institute on Aging, 2019). People who are socially isolated cost Medicare $6.7
billion yearly due to higher service utilization (Flowers et al., 2017). Social isolation increases
the risk of developing dementia by 50%, coronary heart disease by 29%, and stroke by 32%
(Holt-Lunstad, 2020; NASEM, 2020; Valtorta et al., 2016). Research has indicated a link
between social isolation and dementia with lower socioeconomic neighborhoods, race, and
limited social resources, suggesting higher prevalence with these factors (NASEM, 2020;
Samuel et al., 2018; Tigges et al., 1998). The current COVID-19 pandemic has created a new
risk factor for people, who are required to stay at home with severely diminished social contact
(Morrow-Howell et al., 2020).
Study Limitations
Several limitations are important and highlight the need for further research with this
population. First, a change in social engagement or social network interactions may be an early
sign of cognitive impairment as individuals withdraw during the very early stages of the
dementia process (Fratiglioni et al., 2004). Second, individuals categorized as normal at baseline
and having likely dementia at follow-up could have experienced mild cognitive impairment at
baseline given the short follow-up period. Third, the MMSE is a measure used to determine
cognitive status as a screening instrument and assess the need for further neuropsychological
evaluation. Fourth, the study relied on self-report measures for all independent variables, leading
the accuracy of the outcomes to hinge on participants responding with honesty irrespective of
social desirability, no response bias, and clarity about the meaning of the questions (Demetriou et
al., 2014). Last, this study provided insights into the features of social engagement that protect
against cognitive decline using a small sample size, which may cause imprecise estimates (Long
& Long, 1997; Nemes et al., 2009).
54

Policy Implications
Social engagement exists within the constraints of society and the basic infrastructures
that facilitate the maintenance of social engagement. We know from the social trajectory model
that social conditions, including social engagement in adulthood, can change health outcomes
(Berkman et al., 2011). However, many factors affected social engagement, including those
embedded in the social determinants of health paradigm as described by the NASEM (2020)
guiding framework. In other words, individuals exist within the opportunities afforded by their
community and society. The WHO has recognized this with initiatives such as age-friendly cities
and health in all policies. The tenet of age-friendly cities is the ability to enable “people of all
ages to actively participate in community activities and treats everyone with respect, regardless
of their age” (WHO, 2007).
Transforming Health in All Policies to Social in All Policies
Health in all policies is the systematic incorporation of health and health equity into
decision making across sectors and policy areas. The WHO (2013) defined it as “an approach to
public policies across sectors that systematically takes into account the health implications of
decisions, seeks synergies, and avoids harmful health impacts to improve population health and
health equity.” The definition of health in all policies and many of the related efforts that
policymakers and stakeholders have developed have left out social well-being (Holt-Lunstad,
2020). Clear scientific evidence suggests that social infrastructure is critical to the well-being of
individuals, communities, and societies (Carlson et al., 2017; Livingston et al., 2020; NASEM,
2020). Holt-Lunstad (2020) proposed the adoption of a parallel framework, social in all policies.
The goal would be to systematically evaluate existing policies at all levels that interfere with
social engagement but through intersectoral collaboration, which could have far-reaching
55

consequences in areas such as the health and technology sectors. To better understand this
concept and how it could be applied in industry, Table 12 provides examples adapted from Elder
and Retrum (2012) of life factors that may trigger social isolation. The examples are followed by
a discussion of a social in all policies approach in the health and technology sectors. Policies that
support social engagement on face value may not seem to be related to cognitive impairment, but
they are likely to have a spillover effect on dementia prevention.
Table 12. Examples of Potential Triggers Interfering with Social Engagement
Potential Triggers of Isolation Common Examples
Change or loss of social network Death of spouse or close friends and family
Location Rural, inaccessible, or unsafe community setting
Change or Loss of Social Network
A target within reach of Congress is the adoption of policies that seek to incorporate
social isolation screenings during annual wellness visits to assess individuals who may be at high
risk of social isolation and provide primary interventions to those individuals (Flowers et al.,
2017; NASEM, 2020). At present, there is no standard of care practice to address the lack of
social engagement—social isolation (NASEM, 2020). According to the 2020 NASEM report on
social isolation, the current research literature indicates that the adverse health outcomes of
social isolation create increased health services utilization, which provides an opportunity for
screening and primary interventions. The incorporation of social isolation screening during
wellness visits could be perceived as an added burden by health care providers who may have
limited time between patients, and any new policy initiative will need to address this factor
(Holt-Lunstad, 2020; Holt-Lunstad et al., 2017; NASEM, 2020). The incorporation of social
isolation screenings during annual wellness visits has the potential to have an effect at the
56

population level with a system of care that can identify social isolation. The health care system
has the infrastructure necessary to more easily undertake screenings and provide appropriate
education, referrals, and follow-up care. Dementia advocates have pushed for early detection to
possibly delay symptoms (UsAgainstAlzheimer’s, 2021). Identifying individuals with low social
engagement in the health care system opens the opportunity for the provision of intervention
programs.
Health Care System Considerations
Today, we understand health behaviors and health care treatment greatly varies among
cultural groups and by socioeconomic status. As such, Latinos tend to postpone seeking initial
treatment and support in comparison to non-Latino Whites (Lin et al., 2020; Sayegh & Knight,
2013). The burden on the health and human services system and families demands the fine tuning
of approaches to address this public health problem (Betancourt et al., 2003). To overcome these
barriers, public health policy and dementia care experts recommend that (a) health care systems
increase collaboration and create culturally tailored community engagement and education efforts
to promote Alzheimer’s disease literacy and early detection; (b) the health care workforce that is
able to provide culturally and linguistically competent services such as language concordant care
and support for nonmedical needs be increased; (c) additional in-service training be provided to
health care providers to reduce implicit biases related to Latinos’ low education levels, treatment
adherence, and cultural backgrounds; (d) incentives be provided to health care systems to engage
the community in the process to developing, testing, and expanding successful models; and (e)
diversity be increased in clinical trials (Alzheimer’s Association, 2021; Ortega et al., 2015; Wu
et al., 2016).
57

Adverse Location Settings
The coronavirus pandemic has created unsafe communities and disrupted existing social
engagement patterns, resulting in severely diminished social contact (Morrow-Howell et al.,
2020). The United Nations (2020) has called for strengthening “social inclusion and solidarity
during physical distancing.” Reliance on digital access has allowed society to continue
functioning, but older adults, particularly those with low socioeconomic status and in rural areas,
have been left digitally and socially disconnected. In 2019, the White House Office of Science
and Technology Policy released a report that acknowledged social communication technologies
as key to maintaining social engagement (National Science and Technology Council, 2019).
Today, roughly 22 million U.S. older adults lack wired broadband access at home (Older Adults
Technology Services, 2021). Latino older adults are 3.3 times less likely to have internet, and
older adults with less than a high school diploma are about 10 times less likely (National Science
and Technology Council, 2019). An alternative is for the federal government to expand current
subsidies under the Federal Communications Commission’s Emergency Broadband Benefit
Program to focus on people of all ages, clarifying the program is not limited to students. The
program provides a monthly $50 discount toward broadband service and up to $100 for a digital
device (Federal Communications Commission, 2021). The digital divide falls along age, race,
immigration status, functional limitations, and income factors, requiring policymakers and key
stakeholders to create culturally relevant awareness campaigns (Older Adults Technology
Services, 2021). Although expanding the broadband benefit program can help many families, it
is crucial to keep sight of the many older adults making decisions about paying for either
medications or food, necessitating more than discounted rates. Policymakers and key
stakeholders can address infrastructure issues such as increasing or removing data caps,
58

preventing contract terminations, waiving late fees, and opening Wi-Fi hotspots to all (Beacon et
al., 2020).
Intergenerational programs that teach grandchildren, including middle and high school
students, how to teach older adults, a train-the-trainer model, could provide grandchildren with
the necessary knowledge to teach older adults how to use technology. Improving digital literacy
is key for bolstering older adults’ competence and increasing their ability to use digital
technologies. Accessibility among limited English speakers requires additional instruction and
programs that use a stepped approach to building digital skills.
It is critical to recognize that digital access may not be feasible for everyone, and other
low-technology approaches need to be implemented, such as telephone support. The Older
Americans Act funds Title III-B programs that include telephone wellness check phone calls
with the individual’s agreement, known as “telephone reassurance services.” For example, in
California, during the pandemic, services were expanded through additional funds (less than 5%)
from the federal Coronavirus Aid, Relief, and Economic Security Act (McCoy-Wade, 2020). As
policymakers and key stakeholders take stock of technology and the digital divide, the allocation
and use of funds that include a social in all policies approach will help address social inequities
using various strategies across sectors.
The Role of Age-Friendly Cities in Supporting Change
The success of any policy implementation is tied to a successful on-the-ground
implementation of interventions and programs. Age-friendly cities and communities initiatives
have the political will and human capital to implement change. These initiatives outline discrete
recommendations that address livability domains, including social engagement. A social in all
approach would encourage these social participation and inclusion recommendations to be
59

simultaneously implemented with other programs. The policies discussed here could mitigate a
change or loss in network, and adverse location settings can be supported through age-friendly
cities initiatives, for example:
 Initiatives that build out urban environments, including housing and outdoor spaces, can
incorporate features that help older adults access public transportation and provide
designated areas where older adults can safely wait for rides to medical appointments
 Free and secure wireless internet hotspots can be integrated for engagement in
telecommunications
 Physical built environment designs can include physical spaces where residents can
congregate, including redesigns of park spaces that have disability access features to
allow individuals of all ages with mobility issues to utilize these spaces, and support
educational initiatives to promote intergenerational exchanges
Ongoing implementations of these recommendations can take a systematic approach to
evaluating how to inject facilitators of social engagement into other campaigns. In the long run,
strategies that work together to address other inequities in areas such as food, energy, and
education will increase the ability to implement sustainable changes (Buffel et al., 2019).
Conclusion
The research findings are conclusive: The lack of social engagement creates social
isolation that negatively affects people’s health and increases the likelihood of dementia.
Proactive approaches inclusive of all ages can create age-friendly environments that allow
individuals to thrive into old age. The support of social engagement should be considered when
developing strategies and policies, planning, and providing services, regardless of the sector.
This study focused on Mexican Americans known to have strong ties with family and friends,
60

who may be similar to other ethnic groups (Schwartz, 2007), and these findings may be of
relevance beyond the Mexican American population. Future research should focus on
understanding factors that influence dementia among diverse older adults.

61

REFERENCES
AARP Livable Communities. (2021). AARP network of age-friendly states and communities.
https://www.aarp.org/livable-communities/network-age-friendly-communities/info-
2014/member-list.html
Acevedo, A., Loewenstein, D. A., Barker, W. W., Harwood, D. G., Luis, C., Bravo, M., Hurwitz,
D. A., Aguero, H., Greenfield, L., & Duara, R. (2000). Category fluency test: Normative
data for English-and Spanish-speaking elderly. Journal of the International
Neuropsychological Society, 6(7), 760–769.
Achenbaum, W. A. (2009). A metahistorical perspective on theories of aging. In V. L. Bengtson,
D. Gans, N. M. Putney, & M. Silverstein (Eds.), Handbook of theories of aging (2nd ed.,
pp. 25–38). Springer.
Ai, A. L., Plummer, C., Heo, G., Lemieux, C. M., Simon, C. E., Taylor, P., & Copeland, V. C.
(2011). Racial identity–related differential attributions of inadequate responses to
Hurricane Katrina: A social identity perspective. Race and Social Problems, 3(1), 13–24.
https://doi.org/10.1007/s12552-011-9039-1
Aisen, P. S., Cummings, J., Jack, C. R., Morris, J. C., Sperling, R., Frölich, L., Jones, R. W.,
Dowsett, S. A., Matthews, B. R., & Raskin, J. (2017). On the path to 2025:
Understanding the Alzheimer’s disease continuum. Alzheimer’s Research & Therapy,
9(1), 60.
Alley, D., Liebig, P., Pynoos, J., Banerjee, T., & Choi, I. H. (2007). Creating elder-friendly
communities: Preparations for an aging society. Journal of Gerontological Social Work,
49(1–2), 1–18. https://doi.org/10.1300/J083v49n04_01
62

Alzheimer’s Association. (2004). Hispanics/Latinos and Alzheimer’s disease.
http://www.alz.org/national/documents/report_hispanic.pdf
Alzheimer’s Association. (2019). 2019 Alzheimer’s disease facts and figures. Alzheimer’s &
Dementia, 15(3), 321–387.
Alzheimer’s Association. (2020, Mar 10). 2020 Alzheimer’s disease facts and figures.
Alzheimer’s & Dementia, 16(3), 391–460. https://doi.org/10.1002/alz.12068
Alzheimer’s Association. (2021). 2021 Alzheimer’s disease facts and figures.
https://www.alz.org/media/documents/alzheimers-facts-and-figures.pdf
Amano, T., Morrow-Howell, N., & Park, S. (2019). Patterns of social engagement among older
adults with mild cognitive impairment. The Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 75(7), 1361–1371.
Amieva, H., Stoykova, R., Matharan, F., Helmer, C., Antonucci, T. C., & Dartigues, J.-F. (2010).
What aspects of social network are protective for dementia? Not the quantity but the
quality of social interactions is protective up to 15 years later. Psychosomatic Medicine,
72(9), 905–911.
Angel, J. L., Angel, R. J., Aranda, M. P., & Miles, T. P. (2004). Can the family still cope? Social
support and health as determinants of nursing home use in the older Mexican-origin
population. Journal of Aging and Health, 16(3), 338–354.
Angel, R. J., & Angel, J. L. (2009). Hispanic families at risk. Springer.
Angel, R. J., Angel, J. L., & Hill, T. D. (2015). Longer lives, sicker lives? Increased longevity
and extended disability among Mexican-origin elders. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 70(4), 639–649.
63

Ardila, A. (1995). Directions of research in cross-cultural neuropsychology. Journal of Clinical
and Experimental Neuropsychology, 17(1), 143–150.
Artiga, S., & Hinton, E. (2019). Beyond health care: The role of social determinants in
promoting health and health equity. Health, 20, 10.
Bassuk, S. S., Glass, T. A., & Berkman, L. F. (1999). Social disengagement and incident
cognitive decline in community-dwelling elderly persons. Annals of Internal Medicine,
131(3), 165–173.
Baumgart, M., Snyder, H. M., Carrillo, M. C., Fazio, S., Kim, H., & Johns, H. (2015). Summary
of the evidence on modifiable risk factors for cognitive decline and dementia: A
population-based perspective. Alzheimer’s & Dementia, 11(6), 718–726.
Baxter, J., She’Ierly, S. M., Eby, C., Mason, L., Cortese, C. F., & Hamman, R. F. (1998). Social
network factors associated with perceived quality of life: The San Luis Valley Health and
Aging Study. Journal of Aging and Health, 10(3), 287–310.
Beacon, R., Bennett, A., Beverton-Palmer, M., & Yiu, C. (2020). Digital policy for a lockdown:
How tech can help us adapt to a radically altered world. Tony Blair Institute for Global
Change. https://institute.global/policy/digital-policy-lockdown-how-tech-can-help-us-
adapt-radically-altered-world
Berkman, L. F. (2009). Social epidemiology: Social determinants of health in the United States:
are we losing ground? Annual Review of Public Health, 30, 27–41.
Berkman, L. F., Ertel, K. A., & Glymour, M. M. (2011). Aging and social intervention: Life
course perspectives. In L. George (Ed.), Handbook of aging and the social sciences (7th
ed., pp. 337–351). Elsevier.
64

Berkman, L. F., & Glass, T. (2000). Social integration, social networks, social support, and
health. Social Epidemiology, 1, 137–173.
Betancourt, J. R., Green, A. R., Carrillo, J. E., & Ananeh-Firempong, O., II. (2003). Defining
cultural competence: A practical framework for addressing racial/ethnic disparities in
health and health care. Public Health Reports, 118(4), 293–302.
Black, S. A., Espino, D. V., Mahurin, R., Lichtenstein, M. J., Hazuda, H. P., Fabrizio, D., Ray,
L. A., & Markides, K. S. (1999). The influence of noncognitive factors on the Mini-
Mental State Examination in older Mexican-Americans: Findings from the Hispanic
EPESE. Journal of Clinical Epidemiology, 52(11), 1095–1102.
Blazer, D. G., Yaffe, K., & Liverman, C. T. (2015). Cognitive aging: Progress in understanding
and opportunities for action. National Academies Press.
Bondareff, W., Raval, J., Woo, B., Hauser, D. L., & Colletti, P. M. (1990). Magnetic resonance
imaging and the severity of dementia in older adults. Archives of General Psychiatry,
47(1), 47–51.
Bour, A., Rasquin, S., Boreas, A., Limburg, M., & Verhey, F. (2010). How predictive is the
MMSE for cognitive performance after stroke? Journal of Neurology, 257(4), 630–637.
Brookmeyer, R., Gray, S., & Kawas, C. (1998). Projections of Alzheimer’s disease in the United
States and the public health impact of delaying disease onset. American Journal of Public
Health, 88(9), 1337–1342.
Buffel, T., Phillipson, C., & Rémillard-Boilard, S. (2019). Age-friendly cities and communities:
New directions for research and policy. In D. Gu & M. E. Dupre (Eds.), Encyclopedia of
gerontology and population aging. Springer.
65

Campbell, N. L., Unverzagt, F., LaMantia, M. A., Khan, B. A., & Boustani, M. A. (2013). Risk
factors for the progression of mild cognitive impairment to dementia. Clinics in Geriatric
Medicine, 29(4), 873–893.
Cantu, P., & Markides, K. (2019). Hispanic EPESE (Established Population for the
Epidemiological Study of the Elderly). In D. Gu & M. E. Dupre (Eds.), Encyclopedia of
gerontology and population aging. Springer. https://doi.org/10.1007/978-3-319-69892-
2_3-1
Carlson, M., Gow, A. J., Litwin, H., Kim, E., Kondo, K., Qiu, C., & Roos, V. (2017). The brain
and social connectedness: GCBH recommendations on social engagement and brain
health. https://www.aarp.org/content/dam/aarp/health/brain_health/2017/02/gcbh-social-
engagement-report.pdf
Clark, C. M., Sheppard, L., Fillenbaum, G. G., Galasko, D., Morris, J. C., Koss, E., Mohs, R., &
Heyman, A. (1999). Variability in annual Mini-Mental State Examination score in
patients with probable Alzheimer disease: A clinical perspective of data from the
Consortium to Establish a Registry for Alzheimer’s Disease. Archives of Neurology,
56(7), 857–862.
Clark, E., Fredricks, K., Woc-Colburn, L., Bottazzi, M. E., & Weatherhead, J. (2020).
Disproportionate impact of the COVID-19 pandemic on immigrant communities in the
United States. PLOS Neglected Tropical Diseases, 14(7), e0008484.
Cockrell, J. R., & Folstein, M. F. (2002). Mini-Mental State Examination. In J. R. M. Copeland,
M. T. Abou-Saleh, & D. G. Blazer (Eds.), Principles and practice of geriatric psychiatry
(pp. 140–141). Wiley.
66

Crum, R. M., Anthony, J. C., Bassett, S. S., & Folstein, M. F. (1993). Population-based norms
for the Mini-Mental State Examination by age and educational level. JAMA, 269(18),
2386–2391.
Cumming, E., & Henry, W. E. (1961). Growing old: The process of disengagement. Basic
Books.
Dahlgren, G., & Whitehead, M. (1991). European strategies for tackling social inequities in
health: Levelling up part 2. World Health Organization.
DeKosky, S. T., Shih, W.-J., Schmitt, F. A., Coupal, J., & Kirkpatrick, C. (1990). Assessing
utility of single photon emission computed tomography (SPECT) scan in Alzheimer
disease: Correlation with cognitive severity. Alzheimer Disease and Associated
Disorders, 4(1), 14–23.
Dementia Friendly America. (2015). The Dementia Friendly Initiative.
http://www.dfamerica.org/what-is-dfa
Demetriou, C., Ozer, B. U., & Essau, C. A. (2014). Self-report questionnaires. In R. C. Cautin &
S. O. Lilienfeld (Eds.), The encyclopedia of clinical psychology. John Wiley & Sons.
Downer, B., Al Snih, S., Howrey, B. T., Raji, M. A., Markides, K. S., & Ottenbacher, K. J.
(2019). Combined effects of cognitive impairment and pre-frailty on future frailty and
death in older Mexican Americans. Aging & Mental Health, 23(10), 1405–1412.
Downer, B., Garcia, M. A., Raji, M., & Markides, K. S. (2019). Cohort differences in cognitive
impairment and cognitive decline among Mexican-Americans aged 75 years or older.
American Journal of Epidemiology, 188(1), 119–129.
67

Elder, K., & Retrum, J. (2012). Framework for isolation in adults over 50. AARP Foundation.
https://www.aarp.org/content/dam/aarp/aarp_foundation/2012_PDFs/AARP-Foundation-
Isolation-Framework-Report.pdf
Ertel, K. A., Glymour, M. M., & Berkman, L. F. (2008). Effects of social integration on
preserving memory function in a nationally representative US elderly population.
American Journal of Public Health, 98(7), 1215–1220.
Federal Communications Commission. (2021). Emergency broadband benefit.
https://www.fcc.gov/broadbandbenefit
Flowers, L., Houser, A., Noel-Miller, C., Shaw, J., Bhattacharya, J., Schoemaker, L., & Farid, M.
(2017). Medicare spends more on socially isolated older adults. Insight on the Issues,
125, 1119–1143.
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. Journal of Psychiatric
Research, 12(3), 189–198.
Fratiglioni, L., Paillard-Borg, S., & Winblad, B. (2004). An active and socially integrated
lifestyle in late life might protect against dementia. The Lancet Neurology, 3(6), 343–353.
Garcia, M. A., Saenz, J. L., Downer, B., Chiu, C.-T., Rote, S., & Wong, R. (2018). Age of
migration differentials in life expectancy with cognitive impairment: 20-year findings
from the Hispanic-EPESE. The Gerontologist, 58(5), 894–903.
Garcia-Diaz, A. I., Segura, B., Baggio, H. C., Marti, M., Valldeoriola, F., Compta, Y., Vendrell,
P., Bargallo, N., Tolosa, E., & Junque, C. (2014). Structural MRI correlates of the MMSE
and pentagon copying test in Parkinson’s disease. Parkinsonism & Related Disorders,
20(12), 1405–1410.
68

Gaugler, J., James, B., Johnson, T., Marin, A., & Weuve, J. (2019). 2019 Alzheimer’s disease
facts and figures. Alzheimer’s & Dementia, 15(3), 321–387.
Glass, T. A. (2000). Psychosocial intervention. In L. F. Berkman & I. Kawachi (Eds.), Social
epidemiology (pp. 267–305). Oxford University Press.
Global Council on Brain Health. (2020). AARP.
https://www.aarp.org/content/dam/aarp/health/brain_health/2020/08/GCBH_Summary_O
ne_Pager_08_13_2020.pdf
Gonzalez, H. M. (2020). We don’t know enough about Latinos and Alzheimer’s, and that’s a
huge problem. Los Angeles Times. https://www.latimes.com/opinion/story/2020-02-
19/latinos-alzheimers-disease-health-risk-factors
González, H. M., Ceballos, M., Tarraf, W., West, B. T., Bowen, M. E., & Vega, W. A. (2009).
The health of older Mexican Americans in the long run. American Journal of Public
Health, 99(10), 1879–1885.
Grebler, L., Moore, J. W., & Guzman, R. C. (1970). The Mexican American people: The nation’s
second largest minority. Free Press.
Greenfield, E. A., Oberlink, M., Scharlach, A. E., Neal, M. B., & Stafford, P. B. (2015). Age-
friendly community initiatives: Conceptual issues and key questions. Gerontologist,
55(2), 191–198. https://doi.org/10.1093/geront/gnv005
Hachinski, V., & Munoz, D. (2000). Vascular factors in cognitive impairment: Where are we
now? Annals of the New York Academy of Sciences, 903(1), 1–5.
Hale, J. M., Schneider, D. C., Gampe, J., Mehta, N. K., & Myrskylä, M. (2020). Trends in the
risk of cognitive impairment in the United States, 1996–2014. Epidemiology, 31(5), 745–
754.
69

Hansen, M. C., & Aranda, M. P. (2012). Sociocultural influences on mental health service use by
Latino older adults for emotional distress: Exploring the mediating and moderating role
of informal social support. Social Science & Medicine, 75(12), 2134–2142.
Hepp, J. (2016). AARP, Los Angeles city and county partner to make region age-friendly. AARP.
http://states.aarp.org/aarp-los-angeles-city-county-partner-make-region-age-friendly/
Hill, T. D., Angel, J. L., Ellison, C. G., & Angel, R. J. (2005). Religious attendance and
mortality: An 8-year follow-up of older Mexican Americans. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 60(2), S102–S109.
Hill, T. D., Burdette, A. M., Angel, J. L., & Angel, R. J. (2006). Religious attendance and
cognitive functioning among older Mexican Americans. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 61(1), P3–P9.
Hill, T. D., Uchino, B. N., Eckhardt, J. L., & Angel, J. L. (2016). Perceived social support
trajectories and the all-cause mortality risk of older Mexican American women and men.
Research on Aging, 38(3), 374–398.
Hochschild, A. R. (1975). Disengagement theory: A critique and proposal. American
Sociological Review, 40(5), 553–569.
Holt-Lunstad, J. (2020). Social isolation and health. Health Affairs.
https://doi.org/10.1377/hpb20200622.253235
Holt-Lunstad, J., Robles, T. F., & Sbarra, D. A. (2017). Advancing social connection as a public
health priority in the United States. American Psychologist, 72(6), 517–530.
Howrey, B. T., Raji, M. A., Masel, M. M., & Peek, M. K. (2015). Stability in cognitive function
over 18 years: Prevalence and predictors among older Mexican Americans. Current
Alzheimer Research, 12(7), 614–621.
70

Hughes, T. F., Flatt, J. D., Fu, B., Chang, C.-C. H., & Ganguli, M. (2013). Engagement in social
activities and progression from mild to severe cognitive impairment: The MYHAT study.
International Psychogeriatrics, 25(4), 587–595.
Irwin, A., Solar, O., & Vega, J. (2008). Social determinants of health, the United Nations
Commission of. In H. K. Heggenhougen (Ed.), International encyclopedia of public
health (pp. 64–69). Academic Press.
Keefe, S. E., Padilla, A. M., & Carlos, M. L. (1979). The Mexican-American extended family as
an emotional support system. Human Organization, 38(2), 144–152.
Kelly, M. E., Duff, H., Kelly, S., McHugh Power, J. E., Brennan, S., Lawlor, B. A., & Loughrey,
D. G. (2017). The impact of social activities, social networks, social support and social
relationships on the cognitive functioning of healthy older adults: A systematic review.
Systematic Reviews, 6(1), 259. https://doi.org/10.1186/s13643-017-0632-2
Klinenberg, E. (2016). Social isolation, loneliness, and living alone: Identifying the risks for
public health. American Journal of Public Health, 106(5), 786–787.
Korczyn, A. D. (2020). Dementia in the COVID-19 period. Journal of Alzheimer’s Disease,
75(4), 1071–1072.
Kuiper, J. S., Zuidersma, M., Voshaar, R. C. O., Zuidema, S. U., van den Heuvel, E. R., Stolk, R.
P., & Smidt, N. (2015). Social relationships and risk of dementia: A systematic review
and meta-analysis of longitudinal cohort studies. Ageing Research Reviews, 22, 39–57.
Landale, N. S., Oropesa, R. S., & Bradatan, C. (2006). Hispanic families in the United States:
Family structure and process in an era of family change. In M. Tienda & F. Mitchell
(Eds.), Hispanics and the future of America (pp. 138–178). National Academies Press.
71

Langa, K. M. (2018). Cognitive aging, dementia, and the future of an aging population. In M. K.
Majmundar & M. D. Hayward (Eds.), Future directions for the demography of aging:
Proceedings of a workshop (pp. 249–268). National Academies Press.
Lara, E., Martín-María, N., De la Torre-Luque, A., Koyanagi, A., Vancampfort, D., Izquierdo,
A., & Miret, M. (2019). Does loneliness contribute to mild cognitive impairment and
dementia? A systematic review and meta-analysis of longitudinal studies. Ageing
Research Reviews, 52, 7–16.
Law, A., Lee, Y. C., Gorritz, M., & Plouffe, L. (2014). Does switching contraceptive from oral
to a patch or vaginal ring change the likelihood of timely prescription refill?
Contraception, 90(2), 188–194. https://doi.org/10.1016/j.contraception.2014.04.005
Lezak, M. D. (1995). Executive function. In Neuropsychological assessment. Oxford University
Press.
Lichtenberg, F. R. (2017). The impact of biomedical innovation on longevity and health. Nordic
Journal of Health Economics, 5(1), 45–57.
Lin, P. J., Daly, A., Olchanski, N., Cohen, J. T., Neumann, P. J., Faul, J. D., Fillit, H. M., &
Freund, K. M. (2020). Dementia diagnosis disparities by race and ethnicity: Health
services research: Cost of care and implications for intervention. Alzheimer’s &
Dementia, 16, e043183.
Livingston, G., Huntley, J., Sommerlad, A., Ames, D., Ballard, C., Banerjee, S., Brayne, C.,
Burns, A., Cohen-Mansfield, J., & Cooper, C. (2020). Dementia prevention, intervention,
and care: 2020 report of the Lancet Commission. The Lancet, 396(10248), 413–446.
72

Livingston, G., Sommerlad, A., Orgeta, V., Costafreda, S. G., Huntley, J., Ames, D., Ballard, C.,
Banerjee, S., Burns, A., & Cohen-Mansfield, J. (2017). Dementia prevention,
intervention, and care. The Lancet, 390(10113), 2673–2734.
Long, J. S., & Long, J. S. (1997). Regression models for categorical and limited dependent
variables (Vol. 7). Sage.
Markides, K. S. (1993-2011). Longitudinal Study of Mexican American Elderly Health
(5R01AG010939-18; Waves 1 and 7). National Archive on Computerized Data on
Aging.
Markides, K. S. (1999). Hispanic Established Populations for the Epidemiologic Studies of the
Elderly, 1993-1994 (Arizona, California, Colorado, New Mexico, and Texas). Inter-
university Consortium for Political and Social Research.
Markides, K. S., Boldt, J. S., & Ray, L. A. (1986). Sources of helping and intergenerational
solidarity: A three-generations study of Mexican Americans. Journal of Gerontology,
41(4), 506–511.
Markides, K. S., & Eschbach, K. (2011). Hispanic paradox in adult mortality in the United
States. In R. G. Rogers & E. M. Crimmins (Eds.), International handbook of adult
mortality (pp. 227–240). Springer. https://doi.org/10.1007/978-90-481-9996-9_11
Markides, K. S., Peek, M., & Angel, R. (2013). Aging, health, and families in the Hispanic
population. In M. Silverstein & R. Giarrusso (Eds.), Kinship and cohort in an aging
society: From generation to generation (pp. 314–331). Johns Hopkins University Press.
Markides, K. S., Ray, L. A., Angel, R., & Espino, D. V. (2009). Hispanic Established
Populations for the Epidemiologic Study of the Elderly (HEPESE) Wave 5, 2004–2005
73

(Arizona, California, Colorado, New Mexico, and Texas). Inter-university Consortium
for Political and Social Research.
Matallana, D., De Santacruz, C., Cano, C., Reyes, P., Samper-Ternent, R., Markides, K. S.,
Ottenbacher, K. J., & Reyes-Ortiz, C. A. (2011). The relationship between education
level and Mini-Mental State Examination domains among older Mexican Americans.
Journal of Geriatric Psychiatry and Neurology, 24(1), 9–18.
McCoy-Wade, K. (2020). Friendship line CA. California Department of Aging.
https://aging.ca.gov/download.ashx?lE0rcNUV0zbXhkL8SR2D%2Bw%3D%3D
Medina, L., Sabo, S., & Vespa, J. (2020). Living longer: Historical and projected life expectancy
in the United States, 1960 to 2060. U.S. Census Bureau.
Mitchell, A. J. (2017). The Mini-Mental State Examination (MMSE): Update on its diagnostic
accuracy and clinical utility for cognitive disorders. In A. J. Larner (Ed.), Cognitive
screening instruments (pp. 37–48). Springer.
Mizoguchi, N., Walker, L., Trevelyan, E., & Ahmed, B. (2019). The older foreign-born
population in the United States: 2012–2016. U.S. Census Bureau.
Monserud, M. A. (2019). Later-life trajectories of cognitive functioning among married and
widowed older men and women of Mexican origin. Journal of Cross-Cultural
Gerontology, 34(3), 307–324.
Morrow-Howell, N., Galucia, N., & Swinford, E. (2020). Recovering from the COVID-19
pandemic: A focus on older adults. Journal of Aging & Social Policy, 32(4–5), 526–535.
National Academies of Science, Engineering, and Medicine. (2020). Social isolation and
loneliness in older adults: Opportunities for the health care system. National Academies
Press.
74

National Center for Health Statistics. (2018). California.
https://www.cdc.gov/nchs/pressroom/states/california/ca.htm
National Center for Health Statistics. (2019). Health, United States, 2018. Centers for Disease
Control and Prevention.
National Institute on Aging. (2019). Social isolation, loneliness in older people pose health risks.
https://www.nia.nih.gov/news/social-isolation-loneliness-older-people-pose-health-risks
National Science and Technology Council. (2019). Emerging technologies to support an aging
population. Executive Office of the President of the United States.
https://trumpwhitehouse.archives.gov/wp-content/uploads/2019/03/Emerging-Tech-to-
Support-Aging-2019.pdf
Nemes, S., Jonasson, J. M., Genell, A., & Steineck, G. (2009). Bias in odds ratios by logistic
regression modelling and sample size. BMC Medical Research Methodology, 9, 56.
Neugarten, B. L. (1974). Age groups in American society and the rise of the young-old. The
Annals of the American Academy of Political and Social Science, 415(1), 187–198.
Neugarten, B. L. (1996). The young-old and the age-irrelevant society. In The meanings of age:
Selected papers of Bernice L. Neugarten (pp. 47–55). University of Chicago Press.
Noe-Bustamante, L., Flores, A., & Shah, S. (2019). Facts on Hispanics of Mexican origin in the
United States, 2017. https://www.pewresearch.org/hispanic/fact-sheet/u-s-hispanics-
facts-on-mexican-origin-latinos
O’Bryant, S. E., Johnson, L., Reisch, J., Edwards, M., Hall, J., Barber, R., Devous, M. D., Sr.,
Royall, D., & Singh, M. (2013). Risk factors for mild cognitive impairment among
Mexican Americans. Alzheimer’s & Dementia, 9(6), 622–631.e621.
75

Office of Disease Prevention and Health Promotion. (2014). Social determinants across the life
stages. U.S. Department of Health and Human Services.
https://www.healthypeople.gov/2020/leading-health-indicators/2020-lhi-topics/Social-
Determinants/determinants
Office of Extramural Research. (2020). Longitudinal Study of Mexican American Elderly Health.
U.S. Department of Health & Human Services.
https://projectreporter.nih.gov/project_info_description.cfm?aid=10017792&icde=52098
093
Older Adults Technology Services. (2021). Aging connected: Exposing the hidden connectivity
crisis for older adults. https://agingconnected.org/wp-content/uploads/2021/02/Aging-
Connected_Exposing-the-Hidden-Connectivity-Crisis-for-Older-Adults.pdf
Ortega, A. N., Rodriguez, H. P., & Vargas Bustamante, A. (2015). Policy dilemmas in Latino
health care and implementation of the Affordable Care Act. Annual Review of Public
Health, 36, 525–544.
Ortman, J. M., Velkoff, V. A., & Hogan, H. (2014). An aging nation: The older population in the
United States. U.S. Census Bureau.
Parker, C., & Philp, I. (2004). Screening for cognitive impairment among older people in Black
and minority ethnic groups. Age and Ageing, 33(5), 447–452.
Penninkilampi, R., Casey, A.-N., Singh, M. F., & Brodaty, H. (2018). The association between
social engagement, loneliness, and risk of dementia: A systematic review and meta-
analysis. Journal of Alzheimer’s Disease, 66(4), 1619–1633.
76

Petersen, R. C., Smith, G. E., Waring, S. C., Ivnik, R. J., Kokmen, E., & Tangelos, E. G. (1997).
Aging, memory, and mild cognitive impairment. International Psychogeriatrics, 9(S1),
65–69.
Pillai, J. A., & Verghese, J. (2009). Social networks and their role in preventing dementia. Indian
Journal of Psychiatry, 51(Suppl. 1), S22–S28.
Plouffe, L., & Kalache, A. (2010, Sep). Towards global age-friendly cities: Determining urban
features that promote active aging. Journal of Urban Health, 87(5), 733–739.
https://doi.org/10.1007/s11524-010-9466-0
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general
population. Applied Psychological Measurement, 1(3), 385–401.
Raji, M. A., Snih, S., Ray, L. A., Patel, K. V., & Markides, K. S. (2004). Cognitive status and
incident disability in older Mexican Americans: Findings from the Hispanic Established
Population for the Epidemiological Study of the Elderly. Ethnicity and Disease, 14(1),
26–31.
Roberts, A., Ogunwole, S., Blakeslee, L., & Rabe, M. (2016). The population 65 years and older
in the United States.
https://www.census.gov/content/dam/Census/library/publications/2018/acs/ACS-38.pdf
Rodríguez-Galán, M. B., & Falcón, L. M. (2010). Patterns of social activity engagement among
older Hispanics and their relationship to sociodemographic and health variables.
Activities, Adaptation & Aging, 34(4), 251–275.
Rosselli, M., & Ardila, A. (2003). The impact of culture and education on non-verbal
neuropsychological measurements: A critical review. Brain and Cognition, 52(3), 326–
333.
77

Rote, S., Angel, J., & Hinton, L. (2019). Characteristics and consequences of family support in
Latino dementia care. Journal of Cross-Cultural Gerontology, 34(4), 337–354.
Rote, S. M., Angel, J. L., Kim, J., & Markides, K. S. (2021). Dual trajectories of dementia and
social support in the Mexican-origin population. The Gerontologist, 61(3), 374–382.
Sabogal, F., Marín, G., Otero-Sabogal, R., Marín, B. V., & Perez-Stable, E. J. (1987). Hispanic
familism and acculturation: What changes and what doesn’t? Hispanic Journal of
Behavioral Sciences, 9(4), 397–412.
Samuel, K., Alkire, S., Zavaleta, D., Mills, C., & Hammock, J. (2018). Social isolation and its
relationship to multidimensional poverty. Oxford Development Studies, 46(1), 83–97.
Satizabal, C. L., Beiser, A. S., Chouraki, V., Chêne, G., Dufouil, C., & Seshadri, S. (2016).
Incidence of dementia over three decades in the Framingham Heart Study. New England
Journal of Medicine, 374(6), 523–532.
Sayegh, P., & Knight, B. G. (2013). Cross-cultural differences in dementia: The Sociocultural
Health Belief Model. International Psychogeriatrics, 25(4), 517–530.
https://doi.org/10.1017/S104161021200213X
Scharlach, A. E., & Lehning, A. J. (2013). Ageing-friendly communities and social inclusion in
the United States of America. Ageing & Society, 33(1), 110–136.
Schwartz, S. J. (2007). The applicability of familism to diverse ethnic groups: A preliminary
study. The Journal of Social Psychology, 147(2), 101–118.
Seeman, T. E., Miller-Martinez, D. M., Stein Merkin, S., Lachman, M. E., Tun, P. A., &
Karlamangla, A. S. (2011). Histories of social engagement and adult cognition: Midlife in
the U.S. Study. The Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 66B(Suppl. 1), i141–i152. https://doi.org/10.1093/geronb/gbq091
78

Settersten, R. A., Jr., & Mayer, K. U. (1997). The measurement of age, age structuring, and the
life course. Annual Review of Sociology, 23, 233–261.
Shiels, M. S., Almeida, J. S., García-Closas, M., Albert, P. S., Freedman, N. D., & de González,
A. B. (2021). Impact of population growth and aging on estimates of excess US deaths
during the COVID-19 pandemic, March to August 2020. Annals of Internal Medicine,
174(4), 437–443.
Shye, D., Mullooly, J. P., Freeborn, D. K., & Pope, C. R. (1995). Gender differences in the
relationship between social network support and mortality: A longitudinal study of an
elderly cohort. Social Science & Medicine, 41(7), 935–947.
Sommerlad, A., Ruegger, J., Singh-Manoux, A., Lewis, G., & Livingston, G. (2018). Marriage
and risk of dementia: Systematic review and meta-analysis of observational studies.
Journal of Neurology, Neurosurgery & Psychiatry, 89(3), 231–238.
StataCorp, L. (2015). Stata treatment-effects reference manual. Stata Press.
Stern, Y., Arenaza-Urquijo, E. M., Bartrés-Faz, D., Belleville, S., Cantilon, M., Chetelat, G.,
Ewers, M., Franzmeier, N., Kempermann, G., Kremen, W. S., O. Okonkwo, N.
Scarmeas, A. Soldan, C. Udeh-Momoh, M. Valenzuela, P. Vemuri, & E. Vuoksimaa,
(2020). Whitepaper: Defining and investigating cognitive reserve, brain reserve, and
brain maintenance. Alzheimer’s & Dementia, 16(9), 1305–1311.
Strout, K. A., & Howard, E. P. (2012). The six dimensions of wellness and cognition in aging
adults. Journal of Holistic Nursing, 30(3), 195–204.
Szumilas, M. (2010). Explaining odds ratios. Journal of the Canadian Academy of Child and
Adolescent Psychiatry, 19(3), 227–229.
79

Tierney, M. C., Szalai, J. P., Dunn, E., Geslani, D., & McDowell, I. (2000). Prediction of
probable Alzheimer disease in patients with symptoms suggestive of memory
impairment: value of the Mini-Mental State Examination. Archives of Family Medicine,
9(6), 527–532.
Tigges, L. M., Browne, I., & Green, G. P. (1998). Social isolation of the urban poor: Race, class,
and neighborhood effects on social resources. Sociological Quarterly, 39(1), 53–77.
Tombaugh, T. N., & McIntyre, N. J. (1992). The Mini-Mental State Examination: A
comprehensive review. Journal of the American Geriatrics Society, 40(9), 922–935.
Turner, N., & Morken, L. (2016). Better together: A comparative analysis of age-friendly and
dementia friendly communities. https://www.aarp.org/content/dam/aarp/livable-
communities/documents-2016/Better-Together-Research-Report.pdf
U.S. Census Bureau. (2017). The nation’s population is becoming more diverse.
https://www.census.gov/newsroom/press-releases/2017/cb17-100.html
U.S. Census Bureau. (2018a). Projected race and Hispanic origin: Main projections series for
the United States, 2017-2060. https://www2.census.gov/programs-
surveys/popproj/tables/2017/2017-summary-tables/np2017-t4.xlsx
U.S. Census Bureau. (2018b). Hispanic population to reach 111 million by 2060.
https://www.census.gov/library/visualizations/2018/comm/hispanic-projected-pop.html
U.S. Census Bureau. (2019). Comparative demographic estimates 2018: ACS 1-year estimates
comparison profiles.
https://data.census.gov/cedsci/table?q=&t=Age%20and%20Sex&table=CP05&tid=ACS
CP1Y2018.CP05&hidePreview=true&lastDisplayedRow=45
80

United Nations. (2020). Policy brief: The impact of COVID-19 on older persons.
https://www.un.org/development/desa/ageing/wp-
content/uploads/sites/24/2020/05/COVID-Older-persons.pdf
UsAgainstAlzheimer’s. (2021). Early detection.
https://www.usagainstalzheimers.org/learn/early-detection
Valtorta, N. K., Kanaan, M., Gilbody, S., Ronzi, S., & Hanratty, B. (2016). Loneliness and social
isolation as risk factors for coronary heart disease and stroke: Systematic review and
meta-analysis of longitudinal observational studies. Heart, 102(13), 1009–1016.
Vásquez, P. M., Tarraf, W., Doza, A., Marquine, M. J., Perreira, K. M., Schneiderman, N., Zeng,
D., Cai, J., Isasi, C. R., & Daviglus, M. L. (2019). The cross-sectional association of
cognitive stimulation factors and cognitive function among Latino adults in Hispanic
Community Health Study/Study of Latinos (HCHS/SOL). Alzheimer’s & Dementia:
Translational Research & Clinical Interventions, 5, 533–541.
Vega, I. E., Cabrera, L. Y., Wygant, C. M., Velez-Ortiz, D., & Counts, S. E. (2017). Alzheimer’s
disease in the Latino community: Intersection of genetics and social determinants of
health. Journal of Alzheimer’s Disease, 58(4), 979–992.
Vega, W. A. (1990). Hispanic families in the 1980s: A decade of research. Journal of Marriage
and the Family, 52(4), 1015–1024.
Vega, W. A. (1995). The study of Latino families: A point of departure. In R. E. Zambrana (Ed.),
Understanding Latino families: Scholarship, policy and practice (Vol. 2, pp. 3–17). Sage.
Vega, W. A., Lloyd, D. A., Kuo, T., Trejo, L., Wu, S., & Shapiro, M. (2015). The 2015 Los
Angeles healthy aging report. http://roybal.usc.edu/wp-
content/uploads/2016/04/USC_Roybal-LA_HealthyAging.pdf
81

Vespa, J., Armstrong, D. M., & Medina, L. (2018). Demographic turning points for the United
States: Population projections for 2020 to 2060.
https://www.census.gov/library/publications/2018/demo/p25-1144.html
Villa, V. M., & Wallace, S. P. (2020). Latinos and dementia: Prescriptions for policy and
programs that empower older Latinos and their families. In H. Y. Adames & Y. N.
Tazeau (Eds.), Caring for Latinxs with dementia in a globalized world: Behavioral and
psychosocial treatments (pp. 385–401). Springer. https://doi.org/10.1007/978-1-0716-
0132-7_22
Whitfield, K. E., Fillenbaum, G. G., Pieper, C., Albert, M. S., Berkman, L. F., Blazer, D. G.,
Rowe, J. W., & Seeman, T. (2000). The effect of race and health-related factors on
naming and memory: The MacArthur studies of successful aging. Journal of Aging and
Health, 12(1), 69–89.
World Dementia Council. (2020). Defining dementia friendly initiatives.
https://worlddementiacouncil.org/sites/default/files/2020-09/DFIs%20-
%20Paper%201_V14.pdf
World Health Organization. (2002). Active ageing: A policy framework.
https://www.who.int/ageing/publications/active_ageing/en/
World Health Organization. (2007). Global age-friendly cities: A guide.
https://www.who.int/ageing/publications/Global_age_friendly_cities_Guide_English.pdf
World Health Organization. (2012). Dementia: A public health priority.
http://apps.who.int/iris/bitstream/10665/75263/1/9789241564458_eng.pdf
World Health Organization. (2013). Framework and statement: Consultation on the drafts of the
“Health in All Policies Framework for Country Action” for the Conference Statement of
82

8th Global Conference on Health Promotion.
https://www.who.int/healthpromotion/conferences/8gchp/frameworkandstatement/en/
World Health Organization. (2015). World report on ageing and health.
http://apps.who.int/iris/bitstream/10665/186463/1/9789240694811_eng.pdf
World Health Organization. (2019). Social determinants of health.
https://www.who.int/social_determinants/en/
Wu, S., Vega, W. A., Resendez, J., & Jin, H. (2016). Latinos and Alzheimer’s disease: New
numbers behind the crisis. https://roybal.usc.edu/wp-content/uploads/2016/10/Latinos-
and-AD_USC_UsA2-Impact-Report.pdf
Yaffe, K. (2018). Modifiable risk factors and prevention of dementia: What is the latest
evidence? JAMA Internal Medicine, 178(2), 281–282.
Zeki Al Hazzouri, A., Haan, M. N., Kalbfleisch, J. D., Galea, S., Lisabeth, L. D., & Aiello, A. E.
(2011). Life-course socioeconomic position and incidence of dementia and cognitive
impairment without dementia in older Mexican Americans: Results from the Sacramento
Area Latino Study on Aging. American Journal of Epidemiology, 173(10), 1148–1158.
Zunzunegui, M.-V., Alvarado, B. E., Del Ser, T., & Otero, A. (2003). Social networks, social
integration, and social engagement determine cognitive decline in community-dwelling
Spanish older adults. The Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 58(2), S93–S100.

83

APPENDICES
Appendix A. Institutional Review Board Approval

84

Appendix B. STATA Commands

85

86

87

88

89

Appendix C. Logistic Regressions

OR P value [95% CI] OR P value [95% CI] OR P value [95% CI] OR P value [95% CI]
SE Quality Component
Completely satisfied 0.756 0.546 (0.31, 1.87) 0.761 0.556 (0.31, 1.89) 0.807 0.649 (0.32, 2.03) 0.696 0.459 (0.27, 1.81)
Very satisfied 0.977 0.961 (0.39, 2.43) 0.996 0.994 (0.40, 2.49) 1.028 0.953 (0.41, 2.60) 0.876 0.790 (0.33, 2.32)
SE Structural Components
1‐2 children seen monthly 0.556 0.344 (0.17, 1.87) 0.569 0.363 (0.17, 1.92) 0.605 0.425 (0.18, 2.08) 0.532 0.326 (0.15, 1.88)
3+ children seen monthly 0.654 0.179 (0.35, 1.22) 0.643 0.166 (0.34, 1.20) 0.644 0.172 (0.34, 1.21) 0.657 0.209 (0.34, 1.27)
1‐2 close family seen monthly 0.713 0.314 (0.37, 1.38) 0.721 0.330 (0.37, 1.39) 0.692 0.279 (0.36, 1.35) 0.849 0.652 (0.42, 1.73)
3+ close family seen monthly 0.361 0.029* (0.14, 0.90) 0.355 0.027* (0.14, 0.89) 0.361 0.030* (0.14, 0.90) 0.413 0.071 (0.16, 1.08)
1‐2 close friends seen monthly 0.787 0.539 (0.37, 1.69) 0.780 0.524 (0.36, 1.67) 0.775 0.516 (0.36, 1.67) 0.815 0.616 (0.37, 1.81)
3+ close friends seen monthly 1.151 0.733 (0.51, 2.58) 1.132 0.764 (0.50, 2.55) 1.145 0.745 (0.51, 2.60) 1.122 0.791 (0.48, 2.62)
Regular church attendance 1.189 0.652 (0.56, 2.53) 1.197 0.640 (0.56, 2.55) 1.098 0.813 (0.51, 2.38) 1.167 0.711 (0.52, 2.64)
High church attendance 0.684 0.249 (0.36, 1.30) 0.690 0.261 (0.36, 1.32) 0.715 0.314 (0.37, 1.37) 0.761 0.431 (0.39, 1.50)
Has family/friends as neighbors 1.237 0.620 (0.53, 2.87) 1.227 0.634 (0.53, 2.85) 1.184 0.699 (0.50, 2.78) 1.183 0.706 (0.49, 2.83)
Married 1.273 0.576 (0.55, 2.97) 1.277 0.572 (0.55, 2.98) 1.270 0.587 (0.54, 3.00) 1.448 0.417 (0.59, 3.54)
1 other in household 0.904 0.852 (0.31, 2.60) 0.916 0.871 (0.32, 2.65) 0.813 0.709 (0.27, 2.41) 0.735 0.590 (0.24, 2.25)
2+ in household 0.819 0.707 (0.29, 2.31) 0.825 0.717 (0.29, 2.33) 0.691 0.500 (0.24, 2.02) 0.576 0.326 (0.19, 1.73)
SE Functional Components
Has somone to count on, talk to,
feels close to family & friends
1.192 0.650 (0.56, 2.55) 1.175 0.677 (0.55, 2.51) 1.081 0.843 (0.50, 2.32) 0.949 0.900 (0.42, 2.14)
cons 0.457 0.254 (0.12, 1.75)
Health Behaviors
1+ risk health behaviors 1.136 0.671 (0.63, 2.05) 1.383 0.299 (0.75, 2.55) 1.671 0.117 (0.88, 3.17)
cons 0.431 0.229 (0.11, 1.70)
Health Conditions
2+ chronic conditions 2.134 0.013* (1.17, 3.87) 2.579 0.003** (1.37, 4.85)
cons 0.329 0.121 (0.08, 1.34)
Socio demographics
Age 75‐79 0.523 0.063 (0.26, 1.03)
Age 85+ 1.402 0.409 (0.63, 3.13)
Education 5‐8th grade level 0.545 0.084 (0.27, 1.09)
Education 9+ grade level 0.281 0.004** (0.12, 0.67)
cons 0.556 0.447 (0.12, 2.52)
* p<0.05, ** p<0.01,***p<0.001
Model 2Model 3Model 4
Social engagement,
health behaviors, and
health conditions
Social engagement,
health behaviors, health
conditions, and socio‐
demographics
Men (n=314)
Social engagement
Social engagement and
health behaviors
Model 1
90

OR P value [95% CI] OR P value [95% CI] OR P value [95% CI] OR P value [95% CI]
SE Quality Component
Completely satisfied 0.917 0.821 (0.43, 1.95) 0.919 0.826 (0.43, 1.95) 0.975 0.949 (0.45, 2.09) 0.901 0.795 (0.41, 1.99)
Very satisfied 0.939 0.870 (0.44, 2.00) 0.940 0.872 (0.44, 2.00) 0.998 0.995 (0.46, 2.14) 0.924 0.844 (0.42, 2.03)
SE Structural Components
1‐2 children seen monthly 0.623 0.251 (0.28, 1.40) 0.623 0.251 (0.28, 1.40) 0.625 0.255 (0.28, 1.40) 0.717 0.436 (0.31, 1.65)
3+ children seen monthly 0.774 0.342 (0.46, 1.31) 0.776 0.346 (0.46, 1.32) 0.789 0.382 (0.46, 1.34) 0.830 0.506 (0.48, 1.44)
1‐2 close family seen monthly 0.539 0.050* (0.29, 1.00) 0.539 0.050* (0.29, 1.00) 0.521 0.040* (0.28, 0.97) 0.511 0.039* (0.27, 0.97)
3+ close family seen monthly 0.737 0.404 (0.36, 1.51) 0.739 0.409 (0.36, 1.51) 0.719 0.369 (0.35, 1.48) 0.717 0.380 (0.34, 1.51)
1‐2 close friends seen monthly 0.864 0.663 (0.45, 1.67) 0.866 0.668 (0.45, 1.67) 0.880 0.703 (0.46, 1.70) 0.845 0.629 (0.43, 1.67)
3+ close friends seen monthly 0.686 0.352 (0.31, 1.52) 0.687 0.355 (0.31, 1.52) 0.679 0.340 (0.31, 1.50) 0.646 0.292 (0.29, 1.46)
Regular church attendance 1.790 0.110 (0.88, 3.66) 1.789 0.110 (0.88, 3.65) 1.787 0.111 (0.87, 3.65) 1.702 0.157 (0.81, 3.56)
High church attendance 0.962 0.893 (0.55, 1.69) 0.963 0.895 (0.55, 1.70) 0.973 0.925 (0.55, 1.72) 1.006 0.984 (0.56, 1.81)
Has family/friends as neighbors 1.196 0.628 (0.58, 2.46) 1.198 0.624 (0.58, 2.46) 1.150 0.706 (0.56, 2.38) 1.119 0.773 (0.52, 2.40)
Married 0.434 0.013* (0.22, 0.84) 0.433 0.013* (0.22, 0.84) 0.438 0.015* (0.23, 0.85) 0.474 0.032* (0.24, 0.94)
1 other in household 1.202 0.553 (0.65, 2.21) 1.201 0.555 (0.65, 2.21) 1.204 0.550 (0.65, 2.21) 1.298 0.421 (0.69, 2.45)
2+ in household 1.023 0.950 (0.51, 2.06) 1.020 0.955 (0.51, 2.06) 1.029 0.937 (0.51, 2.08) 0.982 0.961 (0.48, 2.03)
SE Functional Components
Has somone to count on, talk to,
feels close to family & friends
1.411 0.304 (0.73, 2.72) 1.411 0.305 (0.73, 2.72) 1.420 0.295 (0.74, 2.74) 1.715 0.122 (0.86, 3.40)
cons 0.283 0.019 (0.10, 0.81)
Health Behaviors
1+ risk health behaviors 0.965 0.902 (0.54, 1.71) 1.001 0.997 (0.56, 1.79) 1.155 0.636 (0.64, 2.10)
cons 0.284 0.020 (0.10, 0.82)
Health Conditions
2+ chronic conditions 1.298 0.308 (0.79, 2.14) 1.289 0.336 (0.77, 2.16)
cons 0.244 0.012 (0.08, 0.73)
Socio demographics
Age 75‐79 0.626 0.116 (0.35, 1.12)
Age 85+ 1.855 0.068 (0.95, 3.61)
Education 5‐8th grade level 0.365 0.001*** (0.20, 0.68)
Education 9+ grade level 0.526 0.053* (0.27, 1.01)
cons 0.389 0.124 (0.12, 1.30)
* p<0.05, ** p<0.01,***p<0.001
Social engagement,
health behaviors, and
health conditions
Social engagement,
health behaviors, health
conditions, and socio‐
demographics
Model 1Model 2Model 3Model 4
Women (n=549)
Social engagement
Social engagement and
health behaviors
91

OR P value [95% CI] OR P value [95% CI] OR P value [95% CI] OR P value [95% CI]
SE Quality Component
Completely satisfied 0.870 0.629 (0.49, 1.53) 0.870 0.629 (0.49, 1.53) 0.936 0.821 (0.53, 1.66) 0.859 0.613 (0.48, 1.55)
Very satisfied 0.963 0.896 (0.55, 1.70) 0.972 0.921 (0.55, 1.72) 1.038 0.898 (0.59, 1.84) 0.964 0.903 (0.53, 1.74)
SE Structural Components
1‐2 friends seen monthly 0.856 0.531 (0.53, 1.39) 0.849 0.511 (0.52, 1.38) 0.863 0.553 (0.53, 1.40) 0.877 0.607 (0.53, 1.45)
3+ friends seen monthly 0.889 0.678 (0.51, 1.55) 0.878 0.647 (0.50, 1.53) 0.874 0.636 (0.50, 1.53) 0.833 0.534 (0.47, 1.48)
1‐2 close family seen monthly 0.610 0.029* (0.39, 0.95) 0.612 0.030* (0.39, 0.95) 0.590 0.021* (0.38, 0.92) 0.633 0.052* (0.40, 1.00)
3+ close family seen monthly 0.581 0.052* (0.34, 1.01) 0.574 0.048* (0.33, 0.99) 0.563 0.041* (0.32, 0.98) 0.603 0.080 (0.34, 1.06)
1‐2 close friends seen monthly 0.587 0.113 (0.30, 1.14) 0.591 0.119 (0.31, 1.15) 0.603 0.134 (0.31, 1.17) 0.623 0.172 (0.32, 1.23)
3+ close friends seen monthly 0.704 0.082 (0.47, 1.05) 0.698 0.075 (0.47, 1.04) 0.711 0.093 (0.48, 1.06) 0.739 0.148 (0.49, 1.11)
Regular church attendance 1.414 0.178 (0.85, 2.34) 1.418 0.175 (0.86, 2.35) 1.394 0.198 (0.84, 2.31) 1.446 0.169 (0.86, 2.44)
High church attendance 0.793 0.270 (0.52, 1.20) 0.797 0.281 (0.53, 1.20) 0.806 0.308 (0.53, 1.22) 0.877 0.548 (0.57, 1.34)
Has family/friends as neighbors 1.182 0.543 (0.69, 2.03) 1.173 0.562 (0.68, 2.01) 1.110 0.705 (0.65, 1.91) 1.065 0.825 (0.61, 1.86)
Married 0.812 0.359 (0.52, 1.27) 0.809 0.352 (0.52, 1.26) 0.818 0.379 (0.52, 1.28) 0.871 0.561 (0.55, 1.39)
1 other in household 1.131 0.639 (0.68, 1.89) 1.138 0.622 (0.68, 1.91) 1.128 0.647 (0.67, 1.89) 1.159 0.590 (0.68, 1.98)
2+ in household 1.035 0.901 (0.60, 1.79) 1.039 0.891 (0.60, 1.79) 1.021 0.942 (0.59, 1.77) 0.948 0.855 (0.54, 1.67)
SE Functional Components
Has somone to count on, talk to,
feels close to family & friends
1.242 0.387 (0.76, 2.03) 1.237 0.394 (0.76, 2.02) 1.222 0.423 (0.75, 2.00) 1.303 0.309 (0.78, 2.17)
cons 0.356 0.011 (0.16, 0.79)
Health Behaviors
1+ risk health behaviors 1.139 0.516 (0.77, 1.69) 1.250 0.274 (0.84, 1.87) 1.436 0.089 (0.95, 2.18)
cons 0.345 0.010 (0.15, 0.77)
Health Conditions
2+ chronic conditions 1.581 0.016** (1.09, 2.30) 1.667 0.009** (1.13, 2.45)
cons 0.270 0.002 (0.12, 0.62)
Socio demographics
Age 75‐79 0.608 0.024* (0.39, 0.94)
Age 85+ 1.658 0.047* (1.01, 2.73)
Education 5‐8th grade level 0.443 0.000*** (0.29, 0.69)
Education 9+ grade level 0.451 0.002** (0.27, 0.74)
cons 0.416 0.056 (0.17, 1.02)
* p<0.05, ** p<0.01,***p<0.001
Social engagement,
health behaviors, and
health conditions
Social engagement,
health behaviors, health
conditions, and socio‐
demographics
Full Sample n=863
Social engagement
Social engagement and
health behaviors
Model 1Model 2Model 3Model 4
92

Sharifian, N., Manly, J. J., Brickman, A. M., & Zahodne, L. B. (2019). Social network characteristics and cognitive functioning in ethnically diverse
older adults: The role of network size and composition. Neuropsychology.

Suthers, K., Kim, J. K., & Crimmins, E. (2003). Life expectancy with cognitive impairment in the older population of the United States. The Journals
of Gerontology Series B: Psychological Sciences and Social Sciences, 58(3), S179‐S186.

Abstract (if available)

Abstract Forecasts of the older adult population indicate that the 85 and older group is expected to increase more than two-fold, with a 123% increase, by 2060. Along with longer life expectancy, there is increased diversity in all racial and ethnic groups. The overall Hispanic and Latino (hereafter Latino) population is expected to outpace all other racial and ethnic groups, growing from 18% in 2016 to 22% in 2035. Latinos live longer (2.5 years), but they do so with higher levels of multiple chronic health conditions, disability, and cognitive impairment. Existing studies suggest social engagement is a dementia protective factor. To answer the research question of what features of social engagement are associated with lower severe cognitive decline, binary logistic regressions were performed using the longitudinal Hispanic Established Populations for the Epidemiologic Study of the Elderly data collected in 2004–05 and 2006–07. Participants at baseline with severe cognitive decline were excluded from the analysis to avoid reverse causality. After participants were classified as having normal cognition or likely dementia based on the 6-year follow-up data, the final analytic sample was N = 863. The odds ratio indicates that likely dementia decreases with the frequency of contact with friends, perceived support from family members, marriage (for women), and higher education levels. Policymakers and practitioners can influence population health through multisystem mechanisms using a “social in all policies” approach. The proactive strategies of social in all policies will create age-friendly environments that allow individuals of all ages to thrive.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

More knowledge, better plans: a study of heterogeneity in dementia prevalence and mid-life cognitive changes

PDF

Predicting cognitive decline and dementia in elderly twins from indicators of early life oral health

PDF

Contributions of dynamic cerebrovascular function to cognitive decline and dementia: development and validation of a novel neuroimaging approach

PDF

Uncovering hidden figures: disparities in dementia burden in an aging America

PDF

Life course implications of adverse childhood experiences: impacts on elder mistreatment, subjective cognitive decline, and caregivers' health

PDF

The economic security of an aging minority population: a profile of Latino baby boomers to inform future retirees

PDF

Exploring end-of-life care experiences and disparities among Hispanic populations in the United States

PDF

Crosscultural differences in dementia diagnosis and care-seeking in Hispanic and non-Hispanic white outpatients

PDF

Latina elected officials in California: a call to action to prepare and pipeline Latinas into the political process

PDF

Health care utilization and spending of the U.S. aging population

PDF

The impact of social capital: a case study on the role of social capital in the restoration and recovery of communities after disasters

PDF

Subclinical carotid atherosclerosis, psychosocial measures, and cognitive function in middle- to older-aged adults

PDF

Three essays on aging, wealth, and housing tenure transitions

PDF

Structure, agency, and the Kuznets Curve: observations and implications for sustainability planning in U.S. cities

PDF

A framework for evaluating urban policy and its impact on social determinants of health (SDoH)

PDF

A better method for measuring housing affordability and the role that affordability played in the mobility outcomes of Latino-immigrants following the Great Recession

PDF

Does collaborative R&D policy work? The effect of collaborative R&D policy on innovative activities of firms in Korea

PDF

Three essays on health, aging, and retirement

PDF

Examining the longitudinal influence of the physical and social environments on social isolation and cognitive health: contextualizing the role of technology

PDF

Property and labor formalization in the age of the sharing economy: Airbnb, housing affordability, and entrepreneurship in Havana

Asset Metadata

Creator Aguilar, Iris (author)

Core Title Social engagement and cognitive decline in Mexican Americans: implications for age-friendly cities

School School of Policy, Planning and Development

Degree Doctor of Policy, Planning & Development

Degree Program Planning and Development,Policy

Degree Conferral Date 2021-08

Publication Date 07/17/2021

Defense Date 04/27/2021

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

Tag age-friendly,aging,cognitive decline,cognitive impairment,dementia,H-EPESE,Hispanics,Latino,OAI-PMH Harvest,social engagement,social isolation

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Aguila, Emma (committee chair), Markides, Kyriakos S. (committee member), Trejo, Laura (committee member), Wilber, Kathleen H. (committee member)

Creator Email iaguilar@usc.edu,irisa_71@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15595907

Unique identifier UC15595907

Legacy Identifier etd-AguilarIri-9762

Document Type Dissertation

Format application/pdf (imt)

Rights Aguilar, Iris

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