THE SOCIAL LIVES OF INFORMAL DEMENTIA CAREGIVERS

A THESIS

SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR THE DEGREE OF MASTER OF ARTS

IN THE GRADUATE SCHOOL OF

TEXAS WOMAN’S UNIVERSITY

DEPARTMENT OF SOCIOLOGY

COLLEGE OF ARTS AND SCIENCES

BY

G.J. HODSON, B.A., G.CERT.

DENTON, TX

AUGUST, 2020

Copyright © by G.J. Hodson 2020

 ACKNOWLEDGEMENTS

First and foremost, this entire undertaking would not exist if it were not for Black

(Feminist/Womanist) Twitter, where I finally learned about intersectionality, womanism, and

not being just another mediocre white kid. I can’t document every single person whose

tweets nudged me to stop eavesdropping and take charge of my social justice education, but

the stand-outs include @Karnythia, @FeministaJones, and @Bad_Dominicana (Mikki

Kendall, Michelle Taylor, and Zahira Kelly, respectively); getting blocked by @thetrudz was

one of the best worst things that has ever happened to me. I humbly thank them, and I will

be sending gratuities as I can (#payblackwomen).

A college degree was rare where I came from, a master’s almost fictional. Thank you

to everyone who did not talk down to me or others, but instead cultivated our learning. As

for what I learned once I got there, my instincts told me that this multidisciplinary approach

would structure my interests best, and it has been more than I dreamed it could be. I had the

fortune to earn a Graduate Certificate in Multicultural Women’s and Gender Studies

alongside the master’s process, and those were the theories that rang true for my

experiences. Speaking of multidisciplinarity, I want to recognize the following TWU faculty

for guiding me toward key components that made this project viable: Profs. Sara Ishii,

Danielle Phillips-Cunningham, Agatha Beins, and Claire Sahlin from MWGS; Prof. Linda

Marshall from Social Work (ret.); and Prof. Sharla Snider in the Department of Professional

Education. Within Sociology, thank you to Prof. Cynthia Cready at University of North

Texas and Prof. Philip Q. Yang at TWU for getting me started in statistics. Special thank you

ii
to the Federation of North Texas Area Universities for facilitating richer and more diverse

learning spaces.

Key logistical steps were rescued my both departments’ administrative staffpersons:

Jillian Morales (MWGS) and Ivy Naude (Sociology); I thank yall for saving me from

erroneous paperwork! Many thanks also to the staff at the Blagg-Huey Library, especially

Shelia Bickle in the archives and Jimmie Lyn Harris, who, sadly, passed away last month after

forty years of showing students around.

Earlier drafts of this thesis received vital theoretical feedback from Michelle

Vaughan, PhD, of Wright State University, important readability notes from Dana Sayre,

MA, and incalculable editing and formatting support from Criss Forshay.

I received an immeasurable amount of professional solidarity and encouragement

from current and recent students, including Elizabeth Gabhart, PhD, of Tarrant County

College, Cassie Withey-Rila of the University of Otago, Natalie Frodin, MS, of University of

North Texas, Beth Fawcett, PhD, Avery Labelson, MA, Petina Powers, Danielle Barber, and

Kay Jae Johns of TWU, and Arielle Webb, MS, of Palo Alto University — I thank each of

you for sharing your journeys with me and letting me share in yours.

Thanks to the judge and to J. Selah Richards for helping me find myself again after

caregiving. Thanks to Julia Martin, for remaining a dear friend long after our commonalities

ended. Other folks cheering this work on include Kathleen Yacharn, Jelly Locklin Rains

(THE!), Heidi Hickman, Gabriel Uhuru, Este Bagato (and Pocket House!), Terri Lynne

Hudson, Abigail Grieg, Chelsea Elliot (and Baby Zab), Beverly Cruël-Garcia, and Jess

Wagstrom. You are each amazing and I am proud to know you. Grad school is, regrettably, a

iii
bit antisocial, so I must quietly recognize some folks whose paths diverged along the way:

KAK, KEK, NEK, SWS, CE, HD, TMJ, LW, RC, AYB, EM, MG, CM in Denton, and CM

in Lubbock. May your paths ahead be clear and calm. I’d also like to thank Melissa Smith-

Wilkerson and the staff of Alzheimer’s Association of North Texas for helping me stay

connected to caregivers. What is a calling? This is a calling, and yall help me feel seen.

Thank you to the committee who made this happen. To Profs. Jim Williams and

Jessica Gullion of TWU and to Prof. Dale Yeatts of UNT: none of you ever told me I

couldn’t do something, yet you each helped my expectations become more reasonable in

unreasonable times. I hope there are more moments of shared recognition in our futures.

To Joni and Ava (and Pixie and Snowy): thank you for being the family I come home

to. I knew there was no one else I’d want alongside me when I wrote a thesis and I knew

there was no one else I’d want to be alongside in an apocalypse, but I had no idea we’d all do

both at the same time! Yall have kept me fed when I should’ve starved, raised my spirits

when bleakness leered, and never let me lose my humanity to the process, no matter how

hard it got. Let us build and realize your dreams next, whatever shape the world takes on.

Caregiving can decimate your spirit, but afterward there is more room for

compassion and understanding to grow than you ever knew possible. This thesis and all

endeavors to follow are dedicated to the memory of my adopted grandparents, E.E. and

M.T. Hodson — Ed and E — who were my cheatcode into the middle class and later set the

examples by which I continue to live: unconditional love, chosen family, and mutual care at

every opportunity. I hope I can continue to honor and share your gifts for as long as it takes,

until every person can feel as loved and secure as I felt when you were both still here.

iv
 ABSTRACT

G.J. HODSON

THE SOCIAL LIVES OF DEMENTIA CAREGIVERS

AUGUST 2020

In the coming decades, there may not be enough professional caregivers to meet the

needs of elderly and disabled persons; family and friends will take on more informal care

roles than ever before, with great social impact, yet the social literature lacks consistency and

breadth. This exploratory, quantitative study organizes the disparate social literature on

caregiving and dementia care and suggests a critical theoretical framework to interpret social

factors more fully. Utilizing the Caregiving in the U.S. dataset from 2015, three hypotheses test

this framework for a dementia care population to identify variable combinations that exert

outsized influence on care outcomes. Findings confirm two hypotheses: people of color tend

to develop dementia far earlier than whites and persons who are marginalized in multiple

ways tend to perform more activities of daily living; these findings demonstrate a role for

intersectionality and queer subjectivity in dementia care analyses and interpretation.

v
 TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ....................................................................................................... ii

ABSTRACT .................................................................................................................................... v

LIST OF TABLES......................................................................................................................... viii

LIST OF FIGURES ...................................................................................................................... ix

Chapter

I. INTRODUCTION .......................................................................................................... 1

Caregiving and Dementia Care ......................................................................... 2

Terminology and Clarifications ........................................................... 7
Problem Statement ............................................................................................. 10
Rationale ............................................................................................................... 10
Structure of the Thesis ....................................................................................... 11

II. LITERATURE REVIEW ............................................................................................... 12

Informal Caregiving: An Overview .................................................................. 12

Nationwide Research Trends and Findings....................................... 13
Caregiving and Inequality ..................................................................... 16
Research Criticisms and Conclusion .................................................. 20
The Dementia Care Dyad .................................................................................. 24
Dementia Care in the Context of Elder Care.................................... 24
The Dementia Care Dyad as a Social System.................................... 32
Social Linkages of the Dementia Care Dyad ..................................... 34
Conclusion.............................................................................................. 41
Theoretical Framework ...................................................................................... 42
Curing vs. Caring ................................................................................... 44
Critical Theories: Queer Subjectivity and Intersectionality ............. 51
Conclusions ............................................................................................ 58
Hypotheses........................................................................................................... 58

vi
III. DATA AND METHODS .............................................................................................. 61

Data ....................................................................................................................... 61
Sample ..................................................................................................... 61
Variables ............................................................................................................... 62
Dependent Variables............................................................................. 62
Independent Variables .......................................................................... 63
Intervening Variables ............................................................................ 64
Data Analysis ....................................................................................................... 65
Bivariate Correlations ........................................................................... 65
Multivariate Linear Regressions .......................................................... 66

IV. RESULTS .......................................................................................................................... 68

Descriptive Statistics ........................................................................................... 68

Hypothesis 1........................................................................................... 70
Bivariate Correlations ......................................................................................... 71
Multivariate Linear Regressions ........................................................................ 74
Hypothesis Two..................................................................................... 74
Hypothesis Three .................................................................................. 76

V. CONCLUSION ................................................................................................................ 80

Summary ............................................................................................................... 80
Discussion ............................................................................................................ 81
Limitations ........................................................................................................... 85
Recommendations .............................................................................................. 86
Conclusion ........................................................................................................... 88

REFERENCES .............................................................................................................................. 90

vii
 LIST OF TABLES

Tables Page

1. Dependent Variables Used in Analysis of Social Aspects of

Dementia Care Dyads ......................................................................................................... 63
2. Independent and Intervening Variables Used in Analysis of
Social Aspects of Dementia Care Dyads .......................................................................... 69
3. Medians for Age of Onset, Controlled for Race ............................................................. 70
4. Cross-Tabulation of Age of Onset (Dementia/Memory Problems) by Race............. 72
5. Correlation Matrix of Select Caregiver Demographics and Outcomes ....................... 73
6. Regression Models Predicting Normative Respondents’
Self-Reported Physical Strain ............................................................................................. 75
7. Regression Models Predicting Marginalized Respondents’ Count of
Activities of Daily Living Supported................................................................................. 77

viii
 LIST OF FIGURES

Figures Page
1. Age of Onset (Dementia/Memory Problems) ................................................................ 71

ix
 CHAPTER I

INTRODUCTION

Along with an increase in life expectancy, many nations have begun to see an

increase in conditions associated with old age, especially dementia. A person who develops

dementia experiences progressive, irreversible cognitive decline without necessarily

recognizing the change. As such, dementia has a deeply social impact, altering the way others

view and interact with the person and creating dependence upon others that only increases

over time. Those who serve persons with dementia — whether family members, friends, or

domestic workers hired for the purpose — are their caregivers. Caregivers may work to

maintain or improve the quality of life of those they serve (whether that service is paid or

unpaid), but when their support cannot stabilize or improve the condition (as is the case for

dementia), it can be hard to reconcile the care role with cultural, economic, and familial

notions of success and vocation. Research on dementia and on caregiving at large show that

outcomes for the caregiver and care recipient are deeply intertwined with social facets of

their lives: status, identity, and personal support, among others (Bruhn & Rebach 2014:15-

29, 70-118). Unfortunately, research into caregiving and dementia care has not necessarily

developed its social facets.

In this exploratory, quantitative study, I review social scholarship of caregivers,

dementia, and dementia care dyads, then test a social inquiry into dementia care while

applying intersectional and queer theories by Crenshaw and Halberstam. This chapter

introduces the core topics, terminology, and approach.

1
CAREGIVING AND DEMENTIA CARE

As humans live longer, particularly in industrial or post-industrial nations (see Bell in

Grusky & Weisshaar 2014:1066-1077), there is usually an unspoken expectation that the

portion of years where a person can be vibrant and independent are also expanded (Lock

2002:33), however this is so far not the case. Publishing with the National Institute on Aging

and the National Institutes for Health, the World Health Organization (2011:6-8) reports

developed nations are but the first to face unprecedented proportion and growth for elderly

and disabled populations. This phenomenon has raised the visibility of persons whose roles

and responsibilities were hidden, presupposed, until recently: the caregiver.

Elders in the U.S. can now live for decades after retirement age, but their health has

not kept pace (Eifert et al. 2016; cf. Halberstam 2005:153). However, caregiving is not only

about elders; over a quarter of people ages 18-65 in the U.S. have a disability or chronic

condition, and this figure is also growing (Scales et al. 2019:21). Patients with conditions that

were once a death sentence (such as cerebral palsy or AIDS) now have life expectancies that

approach those of the general population. Caregiving can also include recovery support for

injury, temporary illness, and surgery. The proverbial broken leg takes 6-8 weeks to recover,

during which time an otherwise healthy adult may rely on others for transportation,

groceries, bathing, and logistics; the recovery time for life-threatening forms of cancer or

from a traumatic brain injury, meanwhile, can last well over a year and forever change family

roles. Care outcomes tend to worsen for caregivers as frequency of tasks and duration

increase (Hunt et al. 2017:18-22) — these factors are also linked to a higher risk of elder

abuse (Schulz & Eden 2016:107-108).

2
 Giving care to the sick, the elderly, and the injured is a significant social behavior

among humans (George & Ferraro 2016; Montgomery & Kosloski 2013; Schulz & Eden

2016:86); this significance has not always been assumed, however (Seaman 2018). Since the

beginning of the Industrial Age, changed standards of health, hygiene, and even violence

have lowered death rates, while biomedical discoveries have enabled life to be prolonged,

particularly near its end (Bruhn & Rebach 2014:20). As (white, able-bodied) middle-class

women — traditionally caregivers of both children and the infirm (Schulz & Eden 2016:50,

53-55) — gradually joined the workforce (Suitor, Gilligan, & Pillemer 2016), so too did

families begin to spread out rather than living geographically close. Urban complexity and

dual-income priorities also led to lower birth rates, leaving aging parents with fewer children

to serve as caregiver; a growing portion of elders have no children at all (Schulz & Eden

2016:61-62).

Estimates range from 17-39 percent of U.S. adults giving some type of informal care

to another adult each year (Fox, Duggan, & Purcell 2013), of which dementia care accounts

for nearly a quarter (Hunt et al. 2017:6); other common ailments including recent surgery,

cancer, and loss of mobility (Hunt et al. 2015:29). The tasks involved in caregiving depend

upon the condition of the care recipient, but tend to be measured in two types: activities of

daily living (ADLs) and instrumental activities of daily living (IADLs; Hunt et al. 2015:8;

Roth, Fredman, & Haley 2015; Schulz & Eden 2016:45). ADLs are those activities that are

necessary to live: feeding oneself, minimal hygiene, or being able to move oneself short

distances. IADLs, on the other hand, are those tasks that are necessary in order to function

in society: obtaining groceries and preparing meals, managing finances and housework, and

3
arranging support. High-need caregivers number over 8.5 million, or 3.5 percent of all adults

over age 20 (Schulz & Eden 2016:50, 306; Vitaliano et al. 2011); care is considered "high-

need" in any case involving assistance with two or more ADLs, cancer, probable dementia,

or once daily care is required for longer than 2 years (Schulz & Eden 2016:74, 82-83). The

endurance of informal caregivers grants a silent economic and logistical reprieve for their

doctors and communities, but not without costs for the caregiver (Bruhn & Rebach

2014:160-164; Schulz & Eden 2016:92-96).

In the U.S., caregiving is predominated by care for elders, who receive most of their

care from spouses or their (middle-aged) children (Schulz & Eden 2016:50, 53-54). The rapid

growth of the U.S. elder population has led to more hiring of home healthcare workers

(Glenn 2010:154-160), a fast-moving corporatization of long-term care facilities (Edelman

2018), and strains in the long-term care insurance industry (Bruhn & Rebach 2014:86;

Gleckman 2019; Kwak & Polivka 2014). Even if all families could afford professional

support or institutionalization (Bruhn & Rebach 2014:86), the industry is not large enough to

meet the care needs of elders (Gaugler et al. 2019:40-41; Redfoot, Feinberg, & Houser 2013)

in the present, let alone meet projections (West et al. 2014). As was done before the

separation of production spheres (Fraser 2016), families and communities are stepping up

informally (i.e., without pay or training — Hunt et al. 2015:42-45) to fill this gap.

The combination of informal and formal care may not be enough (Redfoot et al.

2013; Roth et al. 2015; Schulz & Eden 2016:51-53). Baby Boomers (born 1946-1964), whose

unusual size as a birth cohort is the reason for their name and official Census designation,

are expected to need more care in their later years than the younger population will be able

4
to provide at previously observed care participation rates (Feinberg & Spillman 2019;

Redfoot et al. 2013; Ryan, et al. 2012). This disparity may be a warning to future generations,

or it may be unique to Baby Boomers, about whom care gaps have been predicted since the

earliest Millennials were born (Kane & Kane 1980). Either way, the Boomer care gap is

already emergent and changing the relationship between age and care participation; as of

2014, Millennial adults (born 1980-1996) already represent one quarter of all caregivers and

16 percent of dementia caregivers (Hunt et al. 2017:8) — the highest of any birth cohort on

record of comparable age.

As with the concept of caregiving, our understanding and terminology for dementia

have evolved over the last half-century. Physicians no longer believe that moderate memory

loss is normal for a person at any age; the terms senility and demented have been deprecated in

favor of framing all cognitive decline as some form or other (however advanced) of

dementia. A syndrome rather than one specific disease (Haeusermann 2017), dementia

indicates a gradual degradation of the person’s synapses. Although some rare and externally

traced types of cognitive decline can be reversed (Dwolatzky & Clarfield 2003), the

overwhelming majority are associated with aging and general health, and these are

irreversible (Gaugler et al. 2019:8). Irreversible dementias are collectively called Alzheimer’s

Disease and Related Dementias (ADRD), although Alzheimer’s itself is now a catch-all for

cases of irreversible dementia that do not follow other, more specific pathologies (such as

frontotemporal or Lewy body dementias). Our current understanding of ADRD pathology

indicates that most people who will eventually develop dementia begin synaptic decline as

much as twenty years before symptoms are noticeable (Gaugler et al. 2019:5); once

5
symptoms appear, cognitive decline can affect any or every aspect of a person’s ability to

learn and remember, even altering their personality (Keefe 2017).

Dementia is diagnosed in 10 percent of all U.S. elders over age 65, and early onset

dementia is estimated to affect another 200,000 persons under age 65 (Gaugler et al.

2019:17). The proportion of U.S. elders diagnosed with dementia appears to have plateaued,

yet the numbers continue to grow as Baby Boomers age (Gaugler et al. 2019:19-23).

Researchers note that the trajectory and care of early onset dementia can be dramatically

different from other forms, especially Alzheimer’s (Caceres et al. 2016). There is no cure or

treatment for ADRDs in sight, only medications that can ease the descent (Gaugler et al.

2019:10-11); from diagnosis forward (maybe even earlier), the person with dementia grows

increasingly dependent upon others to function.

Because physical symptoms of dementia tend to come in later stages, the impact of

dementia is overwhelmingly felt in social contexts, where expectations and assumptions

about reality can clash due to memory loss, language difficulties, and erratic behavior on the

part of the person experiencing cognitive decline (Beard, Knauss, & Moyer 2009).

Supporting a person with dementia can entail anything (and, eventually, everything) from

transportation and decision-making to hygiene and feeding (Gaugler et al. 2019:31-33). Care

for persons with dementia is said to be among the most challenging of all caregiving

situations, and accounts for nearly half of all elder care (Gaugler et al. 2019:31-32).

Out of every five dementia caregivers, two report being solo caregivers (Hunt et al.

2017:23), but feelings of isolation are common among others as well (Schulz & Eden

2016:104-105). Studies associate social isolation with higher stress from caregiving (called

6
“caregiver burden”), especially when care lasts longer than 2 years (Schulz & Eden 2016:306;

Vitaliano et al. 2011). Even professional dementia caregivers report higher levels of isolation

than other caregivers (Morgan, Innes, & Kosteniuk 2011), exacerbating the infrastructural

strains discussed above.

Terminology and Clarifications

The following terms will be used in this study:

Caregiving: the act of supporting another person through daily living activities. The

term caregiving became necessary in the latter half of the twentieth century to differentiate it

from caretaking (the care for a place, a thing, or some other object). Although the term is

common in U.S. research (U.K. research often uses carer and dementia advocates prefer care

partner), the parameters of its application are inconsistent and rarely explained, leading to

wide disparities and conflicts between findings; studies that sample only adults who have

provided some amount of care some time in the last year will demonstrate very different

findings from those that include care of children with developmental disabilities but sample

only caregivers who served for six months or more. (Child-rearing specialties sometimes

substitute caregiver for parent since the term is more inclusive, but the raising of children

without illness or disability has its own connotations and fields of research and is

consistently excluded from caregiving research.) For this study, the following terms are used

interchangeably: care/caregiving/giving care/care work.

Caregiver: any person who performs caregiving; often linked to the person(s) for

whom they care through possessive pronouns (e.g., “Ed’s caregiver”). The term remains

controversial outside academia — Bruhn and Rebach report that “as many as 75 percent of

7
family caregivers who provide personal care do not self-identify themselves as caregivers

until personal care and homemaker assistance becomes an economic and psychological

burden” (2014:7) — yet there is evidence that intense caregiving experiences alter caregivers’

self-perception (158-160).

Care Dyad: the pairing of a care recipient and their caregiver (usually primary). As will

be discussed in the literature review, many studies discuss care dyads while focusing their

inquiry or analyses on either the caregiver or the care recipient, not both. The term care dyad

is here utilized in the interest of reifying the social nature of care work.

Formal and Informal (Care): a distinction made between care/givers as to whether care

work is paid (formal) or unpaid (informal), with implications that formal care also involves

training and supervision while informal care comes from family or friends. Efforts to

compensate informal caregivers (Eifert et al. 2016; Glenn 2010:174-175) and inconsistent

parameters for what constitutes “care” (Hunt et al. 2017:6n) are blurring this boundary.

Professional caregivers are outside the scope of this study, so all care that is not otherwise

specified can be presumed to mean informal care.

Primary Caregiver: an informal caregiver who assumes the majority of responsibility in

at-home, day-to-day care, regardless of others’ support or decision-making; if a care recipient

stays in a medical or long-term care facility, family (or other support) legally appoint a

primary (informal) caregiver to coordinate professional services.

Secondary Caregiver: any informal caregiver who is not primary or solo. Although social

support and distribution of roles have received some attention, research has yet to develop

the role of secondary caregivers in dementia or other care contexts.

8
 Care/Support System/Network: the care dyad and all who support its caregiving efforts.

As no standard yet exists, this study uses care as described above and support only to refer to

the direct social support or logistical assistance given to care dyads; similarly, I use network

for only family and social involvement, whereas system connotes active participation and may

including medical and other intervention practitioners.

Elder/Senior/Old: description of persons who are socialized as having an advanced

age. In the U.S., 65 is a common cut-off, beyond which everyone is an elder, a senior, or an old

person. Elder is most common in social research, senior is more popular in marketing, and old

person is considered impersonal/impolite.

Intersectionality: a theoretical paradigm that explores the consequences for persons

who are subjected to multiple social disparities. Rather than merely calculating material

outcomes in a straightforward computation, intersectionality examines the distinct ways

marginalizations interact with one another (often, but not always, amplifying material

outcomes) within a specific context; recognizing the complex nature of discrimination and

systemic oppression, intersectionality rejects transposing sets of intersections from one

scenario to another, instead requiring consideration of relevant lived experiences.

Caregiving is a social phenomenon that reflects political, cultural, and economic

priorities from every level of society (Bruhn & Rebach 2014:5; Schulz & Eden 2016:59-66);

as such, this and other studies utilizing a U.S. sample (even secondarily) must center studies

of the local population and not uncritically conflate it with others. Therefore, only studies on

U.S.-based samples are referenced. In the interest of person-led praxis, the following terms

will be favored over other conventions where possible (i.e., not quoting or citing specific

9
materials): Black or Black American over African American; LGBTQ over LGBT; men and

women over male and female; and people of color over “minorities.”

PROBLEM STATEMENT

The purpose of this study is to explore the complex social interactions that affect

informal dementia caregivers while demonstrating relevant quantitative and sociological

approaches to the topics of dementia and dementia care. This will be accomplished through

a critical literature review, followed by testing hypotheses over some of the socially

determined niches of dementia care. These three hypotheses will be tested using the 2015

wave of National Alliance for Caregiving’s (NAC/AARP) Caregiving in the U.S. via

multivariate linear regression.

RATIONALE

This thesis draws from many disciplines to raise critical questions about how

caregiving has been researched and how future research may be organized, suggesting a

theoretical framework is necessary to connect and interpret dementia care across a fuller

range of social interactions, from the personal to the systemic. This study adds to

sociological literature on caregiving and proposes a useful theoretical framework for

interpreting dementia care. It is the first study to apply intersectional quantitative analysis to

the topic of dementia caregiving, and its findings will inform the work of communities,

medical personnel, and service practitioners in the position of serving dementia caregivers.

Pointing a more social lens on dementia care will also advance our contemporary response

to one of sociology’s fundamental problems: Durkheim’s question of solidarity.

10
STRUCTURE OF THE THESIS

This study introduces and complicates literature and research practices around

caregiving, particularly dementia care. In Chapter 2, a literature review shows how

inequalities in care dynamics and lack of disciplinary grounding have come together to

normalize research that does not adequately address the social factors in dementia care.

Theoretical approaches to filling this research gap are discussed, with two critical theories

(Crenshaw 1991, 1993; Halberstam 2005) selected to advance and interpret social meanings

from the data. In Chapter 3, the dataset and methods are outlined, as well as the data and

variables to be explored in analysis. Chapter 4 reports findings as obtained through

controlled bivariate correlation and multivariate regression. In Chapter 5, implications are

discussed and recommendations offered on future directions for dementia care research

from the social sciences.

11
 CHAPTER II

LITERATURE REVIEW

This critical literature review curates findings from the periphery of several

disciplines to identify weaknesses and anchor dementia caregiving scholarship in a

sociological framework. Research into informal caregiving in the United States is broadly

introduced — including demographic patterns, trends in methodology, and predictions —

followed by a summation of research specific to dementia care and how dementia care dyads

may constitute a unique social system. Multiple theoretical approaches are reviewed for

depth and relevance in developing this connection, with critical theories by Crenshaw and

Halberstam selected to advance the present analysis and three hypotheses introduced for

testing.

As will be elaborated in the Theoretical Framework, social literature has not been

standardized for the topic of caregiving; instead, caregiving is a diffuse topic, often

secondary to relevant biomedical conditions or broad, decontextualized demographics.

While the focus here is caregiving for one condition — dementia — there, too, literature

often lacks the theoretical groundwork to assert a focused body of work; rarely in the

literature (and never with coordinated standards) do studies elucidate the care dyad, its

members, and its support system outside of sweeping biomedical and public health

outcomes.

INFORMAL CAREGIVING: AN OVERVIEW

In this section, the history and literature of informal caregiving are broadly

12
introduced. Despite growing belief that caregiving is innate in humans, the term and its

research paradigm are less than a century old, only recently differentiated from other

responsibilities by improved medical care and shifting family and gender dynamics.

Demographics are presented, with health and income inequalities appearing to affect both

who needs care and who provides it. Research into family dynamics is discussed, although

there is limited qualitative research (and little research of any orientation for non-family

caregivers). Caregiving research trends are reviewed and found to demonstrate problems in

design and generalizability.

Nationwide Research Trends and Findings

Research on caregiving emerged by the 1980s (Apesoa-Varano et al. 2015; Roth et al.

2015; van Durme et al. 2012), with early and ongoing attention to how informal caregiving

can negatively affect a caregiver's physical, emotional, and/or financial wellness (Beard &

Fox 2008; Bruhn & Rebach 2014:161; Seeher 2013). Caregivers may face greater risks than

non-caregivers for stress (Crimmins & Vasunilashorn 2016), isolation (Adler & Mehta 2014;

Musich et al. 2017), depression (Roth et al. 2015), relationship challenges (Schulz & Eden

2016:104), and early mortality (Bruhn & Rebach 2014:162-163; George & Ferraro 2016;

Roth et al. 2013; Schulz & Eden 2016:100), yet caregiving is not without its positive aspects

(Pinquart & Sörensen 2003; Roth et al. 2015). Negative consequences for caregivers appear

to be moderated by the caregiver's age, race, and gender (Bruhn & Rebach 2014:111; Schulz

& Eden 2016:63). Studies link social support for the caregiver to better outcomes for the

care recipient (Kelley, Lewis, & Southwell 2017).

Most informal caregivers come from the care recipient’s immediate family according

13
to availability and expectations of the relationship (Cornwell & Schafer 2015); income may

be a factor as well (Bruhn & Rebach 2014:86). Available spouses are the default caregiver;

selection among adult children varies by the size of the family, genders of the adult children

and the parent, and individual availability or willingness (Davey & Szinovacz 2008; Suitor et

al. 2016). Siblings, grandchildren, and other family members care less frequently, with non-

relatives giving care up to 15 percent of the time (Hunt et al. 2015:20; Nichols et al. 2011).

Attempts to discern patterns in caregiver selection remain scattered, but family role and

having a close relationship with the care recipient seem to overshadow individual personality

traits (Cornwell & Schafer 2015; Davey & Szinovacz 2008; Pillemer & Suitor 2006; Suitor et

al. 2016). Most caregivers see their role as a duty, not an option (Hermanns & Mastel-Smith

2012), yet the 50-60 percent of caregivers who report not having a choice are at higher risk

for adverse outcomes (Schulz & Eden 2016:56, 109). The proportion of care provided by

families has slowly declined since the 1980s (Nichols et al. 2011), leading to explosive growth

in professional and institutional care (Bruhn & Rebach 2014:20; Eifert et al. 2016; Kane &

Kane 1980).

Research into caregivers tends to favor analysis of governmental datasets or

structured, public-interest surveys that report broad patterns from a nationwide view

(Feinberg & Spillman 2019; Kim et al. 2018; Raschick & Intergsoll-Dayton, 2004). These

large-scale studies set the foundation for predictions, trajectories, and policies toward

caregiving from a macrosociological level (Apesoa-Varano et al. 2015; George & Ferraro

2016), but overshadow microsociological studies into the nuances of care dynamics that may

challenge existing theories, methods, and assumptions (Apesoa-Varano et al. 2015; Seaman

14
2018). Condition-centered studies of caregivers, meanwhile, tend to assert the perspectives

of medical practitioners and intervention providers (recent examples include Bailes, Kelley,

& Parker 2016; Cloyes, Hull, & Davis 2018; Gitlin, et al. 2017; Gouin et al. 2016; Zhang &

Lee 2017) instead of care dyads or their families, leading to studies from the nursing field

that attempt to bridge the gap (Bruhn & Rebach 2014:22; Koehly 2017).

Caregiving is not generally regarded as a public policy priority, so efforts to improve

the visibility of caregiving often convey its scope in large dollar figures, usually either the

cost or savings to society (Bruhn & Rebach 2014:185-186; Schulz & Eden 2016:131;

Scommegna 2016). “Cost to society” is an estimate of the economic impact of caregiving by

calculating what income, productivity, or career advancement adults theoretically sacrifice in

a given year to become caregivers (Cook, Snellings, & Cohen 2018; Schulz & Eden

2016:127-131); “savings to society” estimates what cumulative healthcare costs are avoided

by families, governments, or the greater healthcare system as caregivers prevent more

expensive interventions. Both models reduce the concepts of wellness and care to dollar

amounts and may fail to report in a way that is locally relatable by communities and families

(George & Ferraro 2016).

The most common instruments in caregiving research are scales of caregiver burden,

which quantify the stress or intensity of caregiving experiences (Beard et al. 2009; Bruhn &

Rebach 2014:109; Hunt et al. 2015:7-8). Some of these instruments count ADLs/IADLs, the

number of hours spent providing care, or some combination of the two (Bruhn & Rebach

2014:131; Hunt et al. 2015:39-40; Pinquart & Sörensen 2004). Caregiver burden is also a

component of larger constructs, such as the Marwit-Meuser Caregiver Grief Inventory

15
(2002); however, no measure of caregiver burden or experience has yet become an academic

or intervention standard. Collectively, these instruments indicate that burden negatively

affects caregivers’ physical and emotional health (Hunt et al. 2015:7-8) and can lead to

“caregiver burnout” (Hermanns & Mastel-Smith 2012; Nichols et al. 2011; Schulz & Eden

2016:34-35), but unexplored factors outside the care dynamic also play a role (Hunt et al.

2015:12-13; Roth et al. 2015).

Caregivers are notably more vulnerable to negative health outcomes if they lack

social support or coping skills, but both members of the care dyad can suffer if a caregiver’s

subjective assessment of the care situation is poor (Bruhn & Rebach 2014:154-157, 163).

Alternately, higher income, more education, and legal marital status seem to provide a

protective effect against worse health outcomes (George & Ferraro 2016); social and other

forms of support seem to help diminish some of the stressors of dementia care by

improving adaptability (Roth et al. 2015). Caregiver burden and other measures reinforce

that caregiving interacts not only with health status but with socioeconomic status,

perpetuating inequalities.

Caregiving and Inequality

Care dyads are subject to the same social forces that create health disparities. Aging,

disabilities, and caregiving reflect economic and social stratifications that not only compound

but exacerbate one another (Bruhn & Rebach 2014:20, 84-93; George & Ferraro 2016;

Richardson & Brown 2016; Schulz & Eden 2016:61). Notably, not only the health of the

care recipient is at stake, as caregivers and care recipients both demonstrate health

vulnerabilities from care circumstances; these are also mediated by gender, race, and

16
socioeconomic factors (Apesoa-Varano et al 2015; Bruhn & Rebach 2014:87, 116; Cook et

al. 2018). While many of these disparities mirror the patterns of social status and health

generally, caregiving is especially susceptible to social dynamics of gender and age;

furthermore, care dyads interface with race and LGBTQ status in unique, sometimes

counterintuitive ways.

Overall, gender norms center women as primary informal caregivers (Apesoa-Varano

et al. 2015; Friedemann & Buckwalter 2014; Suitor et al. 2016). Daughters and daughters-in-

law perform at least one-third of all caregiving (Hunt et al. 2015:6; Oliker 2011; Schulz &

Eden 2016:55), with wives representing another 8-10 percent (Bruhn & Rebach 2014:87;

Hunt et al. 2015:20). Adult sons who do participate in care are less likely to be married and

yet tend to contribute less intensively than daughters (Davey & Szinovacz 2008; Friedemann

& Buckwalter 2014); husbands who take on caregiving late in life (perhaps for the first time)

commonly report a loss of personal identity that may be linked with their perceived gender

roles (Hermanns & Mastel-Smith 2012). Even heterosexual dual-income families depend on

women to perform most care responsibilities (Bruhn & Rebach 2014:22-23; Suitor et al.

2016). The younger a woman is when she first becomes a caregiver, the greater her risk for

poverty later in life (Bruhn & Rebach 2014:10, 87).

Historically, marriage, divorce, and complex blended families have affected caregiver

selection and involvement (Piercy 2007; Schulz & Eden 2016:62-63); in particular, care dyads

who are married to one another may gain positive effects (Bruhn & Rebach 2014:161).

LGBTQ adults seem to give care at a higher rate than non-LGBTQ adults — also

demonstrating equal engagement by men and women (Hunt et al 2015:81) — yet are erased

17
when care research is structured around the nuclear family and other cisheteronormative

assumptions (Halberstam 2005:10; Oliker 2011). This gap may exclude 1 in 11 caregivers

(Hunt et al. 2015:14) and disproportionately affect young adult caregivers (Flinn 2018;

Hodson forthcoming); Millennials — the adult cohort most likely to come from blended

families, to identify as multiracial, and to identify as LGBTQ — are taking on caregiving

early and often (Flinn 2018; Fry 2017). Research has found no significant difference in the

outcomes of care dyads based on the sex, marital status, or LGBTQ identification of

caregivers.

People in general are becoming caregivers earlier in life than in decades past (Bruhn

& Rebach 2014:9; Moon & Dilworth-Anderson 2015), reflecting perhaps the smaller sizes of

contemporary families, a growing sense of idealism and service among young people (Cook

et al. 2018; Pew 2015), and a relaxing of traditional gender norms for men. Younger

caregivers, though beleaguered and less prepared (Brodaty & Donkin 2009), show greater

physical resilience against care burden than older caregivers (Hunt et al. 2015:51, 53; Schulz

& Eden 2016:101), even in cases of dementia (Hunt et al. 2017:21); however, there is

concern that people born after 1980 (Millennials and Generation Z) will devote more years

to caregiving across their lifespans than any previous birth cohorts (Cook et al. 2018; Flinn

2018). Older caregivers, especially those over 65 (who are more expected and expecting to

give care; Scommegna 2016; Suitor et al. 2016), tend to face challenges that are more

financial (Eifert et al. 2016) and physical (Pinquart & Sörensen 2003). The risk of early death

is reportedly greater for older caregivers than comparable non-caregivers (Schulz & Eden

2016:100), although this finding has been recently disputed (Roth et al. 2015). Caregivers of

18
“middle” age (more or less Generation X) may have the best combination of resources, but

also carry the greatest number of responsibilities, not least the likelihood of working full

time, raising children, or even both (Berry, Apesoa-Varano, & Gomez 2015; Bruhn &

Rebach 2014:4, 38).

The relationship between health disparities and care disparities is most evident

among people of color, who (particularly Black men) tend to experience major illness earlier

and more severely than other races (Bruhn & Rebach 2014:99). More specifically, people of

color tend to demonstrate more comorbid conditions, greater ADL needs, and earlier onset

of dementia when compared to whites (Angel, Mudrazija, & Benson 2015; Schulz & Eden

2016:61, 110). Rodriguez (2018) suggests that premature death amongst marginalized

communities exacerbates disparities in later life. Several studies point to people of color

being more resilient against social isolation and caregiver burden than whites (Schulz &

Eden 2016:113), but critics note these studies may lack sufficient controls for other

intervening socioeconomic factors, which can have a multiplier effect on health and life

expectancy (George & Ferraro 2016). Caregivers of color, who are less likely to care alone

and more likely to involve non-relatives than whites (Hunt et al. 2015:20; Schulz & Eden

2016:21), generally reporting more positive and fewer negative consequences to care (Angel

et al. 2015; Schulz & Eden 2016:110); this is especially the case with Black caregivers

(Apesoa-Varano et al. 2015). White caregivers tend to be older, more likely to be supported

by family members, and are less likely to be employed (or more likely to be retired) than

caregivers of color (Bruhn & Rebach 2014:105, 107).

19
 Caregiving as a topic of research is neither broad nor deep, but demographic

projections for the U.S. are increasingly urgent. Family and friends make up most informal

caregivers, yet professional caregivers are doing more than ever before. Large-scale studies

have helped define and publicize the scope of caregiving, yet they omit broader support

systems. Inequalities of broader society are reflected and amplified by caregiving, but

complications also exist. Women give the most care and seem to be the most vulnerable to

its sacrifices, yet people of color seem uniquely protected against some of its worst

consequences. Younger caregivers (i.e., Millennials) are participating in care earlier and more

intensively than previous generations at their age, shifting norms and demonstrating diversity

that appear poised to challenge longstanding assumptions about care dynamics.

Research Criticisms and Conclusion

Caregiving can be a difficult research topic to syncretize, spread as it is across many

academic fields and approaches. Care research has no home discipline or set of methods

around which to develop; often, it exists only as a niche within literature on the care

recipient’s condition (van Durme et al. 2012). Broad studies of caregivers, on the other hand,

frequently favor conventional or convenient designs that may not fit. Scholars have noted

such weaknesses in care research as narrow sampling (Apesoa-Varano et al. 2015; George &

Ferraro 2016; Pinquart & Sörensen 2003), recording and analysis that oversimplify U.S. racial

dynamics (Angel et al. 2015; Bruhn & Rebach 2014:110), and reporting statistics for non-

probabilistic samples (Pinquart & Sörensen 2004; Schulz & Eden 2016:60; Werner, Karnieli-

Miller, & Eidelman 2013). Methods get mimicked uncritically, allowing a single facet of a

care recipient (such as their age, healthcare, or disease) to overshadow potentially meaningful

20
dynamics from their entire support network (Davey & Szinovacz 2008). Advocates are

calling for more inclusion of the care recipient’s perspective in care research (Schulz & Eden

2016:220-221), even in the case of dementia (Fazio et al. 2018).

Although public and public interest datasets convey the scale of informal care by

counting and categorizing caregivers, these datasets only fulfill bureaucratic priorities if

explored uncritically, overlooking even big-picture intimacies of caregiving. Surveys and

interviews alike prefer to examine caregiving through a single unit of analysis — the primary

caregiver — despite evidence that larger care networks are present and effective (Nichols et

al. 2011; Schulz & Eden 2016:106). Critics note that surveying a single caregiver ignores

social networks beyond the care dyad (Davey & Szinovacz 2008; Koehly 2017); the level of

involvement by non-primary caregivers, the sharing of responsibilities, and the selection of a

primary from several options have all been recommended for further inquiry. Examining the

entire phenomenon of caregiving can be challenging because of just how unseen caregiving

tends to be (Seaman 2018). Research designs that emphasize one caregiver and one care

recipient dealing with one condition may not adequately reflect facts on the ground and risk

furthering this invisibility. Isolation and time limitations also can make care dyads difficult to

recruit, especially for lengthy interviews or observation (Bédard et al. 2001; van Durme et al.

2012), and many families from low-income backgrounds — and families of color in general

— reject the construct of “caregiving” entirely (Apesoa-Varano et al. 2015; Bruhn & Rebach

2014:87, 100-105). Deeper still are the limitations of care research grounded in superficial

assumptions and care ontology; studies that focus on “family care” or “elder care” or

“informal care” do not necessarily define their terms consistently, or at all.

21
 George and Ferraro (2015) invoke “the complexity of change” to endorse social

aging research with “considerable audacity and fortitude,” but care scholarship lacks diversity

in both designs and cases (Remler & van Ryzin 2011; Schulz & Eden 2016:56). While the

over-65 population of the U.S. remains overwhelmingly white, the proportion of whites in

this segment is decreasing (Angel et al. 2015; Eifert et al. 2016; combined with lower life

expectancies for people of color, the concentration of racial diversity in people born after

1960 (Schulz & Eden 2016:59-60) risks functional segregation of age itself. This racialization

of U.S. aging may already be underway in formal care, where 26 percent of professionals are

immigrants, who are vulnerable to misinterpretation, exploitation, and mutual distrust (Scales

et al. 2020). Reliance on professional caregivers nearly doubled from 2008 to 2018, and may

continue to do so (Scales et al. 2020); demographers warn that the U.S. population has not

grown fast enough from births alone to meet the anticipated care needs (formal or informal)

of aging Baby Boomers (Kane & Kane 1980; Schulz & Eden 2016:185; Scommegna 2016).

Both quantitative and qualitative approaches risk repeating societal erasures and

diminish academic rigor when they look at gender, race, and class as separate phenomena

from health status, quality of life, or caregiving outcomes (Kelley et al. 2017; Lovell &

Wetherell 2011; Oliker 2011). Increasingly, health in old age is being understood not as a

random occurrence but as the culmination of a person’s life experiences and their

socioeconomic advantages and disadvantages (Richardson & Brown 2016), but this approach

is not often employed in studies of elder care. Angel et al. caution that caregiving surveys do

not adequately address "the way in which social aspects of racial and ethnic diversity are

affected by the life course and social policy," connecting this oversight to poor clarity of

22
what is meant by "social support" (2015:8; cf. Cornwell & Schafer 2015). Without far more

rigorous analysis of many social factors and how they interact with caregiving, Roth et al.

warn that caregiving for a sick relative cannot be demonstrated to be any more stressful than

having a sick relative without providing them care (2015:310-11) — except, as is common

with such critiques, in cases of dementia care (cf. Pinquart & Sörensen 2003).

A growing number of health researchers are calling for multi-level analyses that

target health inequalities (Gkiouleka et al. 2018; Richardson & Brown 2016; Richman &

Zucker 2019). Studies commonly differentiate “high need” care situations — including

dementia (Schulz & Eden 2016:91) — for having stronger feelings of exhaustion,

overwhelm, and burnout (Gaugler et al. 2019:34-36; Pinquart & Sörensen 2003; Schulz &

Eden 2016:74), but the “high need” designation lacks consistent standards. The grounding

of caregiver studies with theory and/or a dedicated discipline may offer better frameworks

around care research, built on nuanced categories (Angel et al. 2015; George & Ferraro

2016) that will, in turn, better inform care dyads and those who support them.

Caregiving scholarship draws from a wide range of traditions, which can lead to

questionable methods and results that are difficult to generalize. Scholars of caregiving and

health have an opportunity to fill the gap of social meanings, particularly the interrelatedness

of social inequalities and clearer understandings of social support, but to do so will require

streamlining diffuse design approaches. At the same time, informal care research must

expand its scope to include not only caregivers but dyads and support networks, not only

health and burden measures but a layering of positive and negative socioeconomic factors

and their systemic roots over the life course. Care research has an urgent need to advance, as

23
generational changes are beginning to outpace foundational assumptions and nationwide

care gaps loom; the scale of care likely to be needed in the coming decades will have a

dramatic effect on the attitudes and opportunities of Millennials and Generation Z over their

lifespans.

THE DEMENTIA CARE DYAD

In this section, dementia care dyads are explored as a unique and significant social

system. Dementia care and its scholarship are outlined relative to other care dyads, followed

by some exploration of research on how this system interacts with other social systems.

Outcomes from these interactions are summarized and trends are detailed.

Dementia Care in the Context of Elder Care

Caregiving research is often framed by the care recipient's condition, and dementia

care is considered one of the most intensive and challenging forms of caregiving (Bruhn &

Rebach 2014:131). Early caregiving research framed family members as “the ‘real’ victims”

of dementia (Beard et al. 2009:227), and even today, “The most prevalent discourses in the

US depict family caregiving as a necessary, but onerous stopgap measure until a better,

biomedical solution is available” (Seaman 2018:64; cf. Cockerham 2013; Mol 2002:7-13n).

Dementia caregivers report high levels of burnout (Bookman & Kimbrel 2011), as well as

difficulties receiving adequate support — social or otherwise (Hunt et al. 2017:25) — yet no

current or near-future treatment for dementia can significantly lower the need for people

who monitor and protect people with dementia (Stites, Rubright, & Karlawish 2018). In

total, nearly 6 million people in the U.S. are living with Alzheimer’s or another dementia and

over three-quarters will need daily care by the late stages (Gaugler 2019:17; Scales et al.

24
2020). Although the proportion of elders experiencing dementia appears to have plateaued,

the population of people with dementia continues to grow (Gaugler et al. 2019:23;

Scommegna 2016).

Caregiving is generally considered "high-need" if daily care lasts for longer than two

years, but dementia care becomes difficult much sooner and lasts an average of 4-6 years

(Gaugler et al. 2019:34-36; Pinquart & Sörensen 2003; Vitaliano et al. 2011). Caregiving for a

person with dementia never stabilizes for long, but gradually expands in scope until the care

recipient loses mobility or dies. Both the intensity and the duration of care increase the

likelihood of negative outcomes for the 16 million dementia caregivers in the U.S. (Gaugler

et al 2019; Bruhn & Rebach 2014:131; Seltzer & Li 1996); further, 66 percent of caregivers

cohabitate with their care recipients, more than any other form of care (Gaugler et al.

2019:31; Pinquart & Sörensen 2003), leaving them with little reprieve. Dementia care

demonstrates negative effects on measures of caregiver wellness in both prevalence and

severity (Roth et al. 2013; Pinquart & Sörensen 2004; Seeher 2013), and the “functional

limitations and behavioral symptoms” associated with dementia (Schulz & Eden 2016:85) are

noted to affect subjective well-being more than high counts of care needs (Schulz & Eden

2016:103; Pinquart & Sörensen 2004). A sense of guilt over what more can or should be

done is common, if usually unwarranted (Bruhn & Rebach 2014:163; Seaman 2018).

The risks of harm from isolation, depression, burnout, sleep deprivation, and

worsened health are higher for dementia caregivers than for other caregivers (Bailes et al.

2016; Bookman & Kimbrel 2011; Greenwood, Mezey, & Smith 2018), particularly for adult

children and grandchildren (Nichols et al. 2011); even formal caregivers working with people

25
with dementia report higher levels of stress and depression than professionals caring for

other conditions (Morgan et al. 2011). Burden measures often come in higher for dementia

caregivers than for most other forms of care (Bailes et al. 2016; Hunt et al. 2017:18) and are

sometimes customized for dementia care (reviewed in Nichols et al. 2011), but it is

important to assess these figures relative to the progression of the disease, since the mid- to

late-stages are associated with worse behavioral problems and a higher intensity of care

overall (Hirschman & Hodgson 2018).

A strong connection between the dyad often predates informal caregiving (Bruhn &

Rebach 2014:10; Suitor et al. 2016), yet caregiver assignment can be more accurately

predicted by the age or condition of the prospective care recipient than by the prospective

caregiver’s own circumstances (Bruhn & Rebach 2014:10). Dementia caregivers describe

their role as an act of love and familial commitment (Dunham & Cannon 2008), but this

attitude is more pronounced in women and families of color than with men and whites

(Bruhn & Rebach 2014:100-105; Shurgot & Knight 2005). Dementia caregivers appear more

likely to themselves have a chronic illness or disability than other caregivers (Bruhn &

Rebach 2014:153-154; Eifert et al. 2016; Wang, Robinson, & Carter-Harris 2014); however,

this connection has not been explored in the literature. Forty percent of dementia caregivers

report being sole caregivers (Hunt et al. 2017:23), but feelings of isolation are common

among others as well (Schulz & Eden 2016:104-105).

Dementia care tends to begin with immediate family, falling to spouses first and then

adult children (Gaugler et al. 2019:31; Nichols et al. 2011), although siblings and

grandchildren are more involved in dementia care than other informal caregiving (Nichols et

26
al. 2011). Some families must change caregivers at least once, but studies are rare (Davey &

Szinovacz 2008); wives, daughters, and siblings seem to be at greater risk of burnout than

husbands, sons, and distant relatives (Nichols et al. 2011:29), yet spouses of persons with

dementia may increase their own risk of developing dementia by giving care (Vitaliano et al.

2011). Most comparative studies have only compared dyads of one relationship type

(spouses, for example) against another (adult children, often daughters), with units selected

across — as opposed to within or beyond — families (Nichols et al. 2011; Schulz & Eden

2016:106; Seltzer & Li 2000).

The disparity between men and women’s prevalence as caregivers is stronger for

dementia care than for other forms of care; wives, daughters, and even daughters-in-law

together perform over two-thirds of all dementia care (Gaugler et al. 2019:31; Friedemann &

Buckwalter 2014; Nichols et al. 2011), though not without “frustrations” over family men

who expect to make decisions but not to participate in hands-on care (Dunham & Cannon

2008:48). Despite some qualitative variations in how care is performed, research into the

gender of dementia caregivers has reported no significant difference in health outcomes for

the care recipient.

Negative caregiver outcomes are often moderated by race or ethnicity (Angel et al.

2015). As with caregiving broadly, families of color show better resilience against negative

care outcomes than white families, but also a stronger aversion to the core research concepts

of “caregiving”, “care burden”, and “dementia” itself (Apesoa-Varano et al. 2015; Bruhn &

Rebach 2014:102; Schulz & Eden 2016:22). Whites are the only racial group to care more

27
often for spouses than parents (Bruhn & Rebach 2014:105, 107), while people of color care

more often than whites for other relatives and non-relatives (Nichols et al. 2011).

Marital status has a strong effect on dyad makeup from both sides. Spouses make up

12 percent of dementia care dyads; outside of spousal dyads, dementia caregivers are more

likely to be married than other caregivers or the general populace (Hunt et al. 2017:35), and

care recipients who are unmarried rely more heavily on siblings, grandchildren, and in-laws

(Nichols et al. 2011). When caring for someone other than a spouse, however, dementia care

has been found to create a greater strain on marriages than other forms of care (Schulz &

Eden 2016:104). Millennial caregivers, who demonstrate the lowest marriage rates on record

and are more likely to identify as LGBTQ, represent a quarter of all dementia caregivers

(Flinn 2018; Gurrentz 2018).

Challenges can vary according to the age dynamic of any care dyad, and dementia

exhibits strong differences by age and age difference. Dementia care seems to face more

severe financial challenges when the caregiver is aged 18-35 (Flinn 2018); their careers,

educations, and resilience against economic oppression may be less developed compared to

both past cohorts of the same age range and older family members of the present, although

there is also evidence that Millennials face greater economic hardship than any other

generation since World War II (Fry 2017). Grandchild caregivers seem to be especially

vulnerable to social isolation, and commonly experience both role conflict and

unpreparedness (Bruhn & Rebach 2014:153; Nichols et al. 2011). Younger caregivers show a

greater physical resilience against care burden, even in cases of dementia (Hunt et al.

2017:21), but Millennials and Generation Z face a ballooning prevalence of care that could

28
overwhelm other expectations about family and work over the life course (Cook et al. 2018;

Eifert et al. 2016). Older caregivers also demonstrate financial strain but report the greatest

impact to their health (Lovell & Wetherell 2011). Just over a third of dementia caregivers are

themselves over 65 (Gaugler et al. 2019:31); some argue dementia care can increase the

mortality of older caregivers, especially spouses (Bruhn & Rebach 2014:162-163; George &

Ferraro 2016; Roth et al. 2013). Dementia caregivers of the Baby Boomer cohort report

higher levels of responsibility and greater disruptions to physical and emotional health

(Lovell & Wetherell 2011; Moon & Dilworth-Anderson 2015; Musich et al. 2017). Middle-

aged caregivers describe not only strong emotional challenges but also the greatest conflicts

of role and responsibility (Bookman & Kimbrel 2011). Caregivers supporting an elder while

also parenting children under 18 are known as the “sandwich generation” (Berry et al.

2015:107) and have a high likelihood of being employed in addition to dual-care

responsibilities (Hunt et al. 2015:56).

Lower income and less education attainment are each associated with better care

outcomes for adult children, though researchers also associate lower income with a stronger

sense of family duty (Bruhn & Rebach 2014:87); neither of these patterns has been explored

deeply in the context of dementia care. Willert and Minnotte note that while research into

caregiving consistently finds caregivers of color or of lower socioeconomic status to be more

vulnerable to certain strains, race and income remain under-examined (2019).

Dementia research exhibits many of the research conventions discussed in the

previous section, such as interviewing the caregiver while treating the care recipient as a

medical case rather than examining the dyad holistically (Seaman 2018). Research into

29
dementia caregivers focuses on biomedical outcomes, utilization of social services, and intra-

family dynamics (Davey & Szinovacz 2008; Schulz & Eden 2016:106), centering not only the

pathological interpretation of dementia but also the so-called nuclear family over other care

dynamics (Apesoa-Varano et al. 2015; Nichols et al. 2011); this narrow view not only

excludes over a third of contemporary caregivers who are unmarried or not caring for

immediate family, it also centers middle class white families in a way that is outdated,

increasingly irrelevant, and methodologically specious (Eifert et al. 2016). The “cost/savings

to society” paradigm is less widely utilized because dementia is seen (somewhat inaccurately)

as an affliction of people who have aged out of the workforce; indeed a quarter of dementia

caregivers are over age 65, and yet 21 percent of these older caregivers still work (Hunt et al.

2015:56; Hunt et al. 2017:8).

Dementia care research often happens at the local or state level, but such studies can

be difficult to place in conversation with one another, or with national studies, due to

methodological inconsistencies. Publicly available datasets such as the federal National Survey

of Caregivers and NAC/AARP’s Caregiving in the U.S. record and report which caregiving cases

center on dementia, but researchers must be careful working with the smaller samples, where

convention favors homogeneous samples that hinder reliability (George & Ferraro 2016).

Even studies that measure biomedical factors upon social phenomena stand accused of poor

sampling (Roth et al. 2015). Perhaps the greatest weakness in dementia care research is the

high number of caregiving studies that lack probability samples (Angel et al. 2015; Pinquart

& Sörensen 2003, 2004; Schulz & Eden 2016:60; Werner et al. 2013), and yet go on to report

statistics probabilistically (such as Wang et al. 2014). The abundance of unstandardized, by-

30
the-numbers statistical methods may hide the opportunity for more innovation and breadth

in care study designs.

Assumptions that caregiving carries a cumulative harm for the caregiver also persists

in dementia care research, which then focuses on quantifying the harm rather than

complicating it (Dunham & Cannon 2008; Hermanns & Mastel-Smith 2012; Roth et al.

2015). Studies that highlight "positive aspects of caregiving" tend to make an explicit

exception for dementia care (such as Roth et al. 2015; Pinquart & Sörensen 2003), although

such studies exist for dementia care as well (Beard et al. 2009). Though, pathologically,

dementia is an individual phenomenon and in no way contagious, the diagnosis of dementia

can shake up entire families and communities. Some of the social and demographic trends

for dementia caregiving follow those of caregiving in general, but others are amplified. In

particular, the role of spouses and daughters is more prominent, and age differences (both

between different caregivers and between a caregiver and their care recipient) have stronger

impact. Caregivers of color demonstrate resilience in dementia care, but their coverage in

literature is inconsistent and often lacks nuance. Specialized burden measures exist for

dementia caregivers, but quantitative studies do not tend to record stages of the disorder in

ways that would make the measures reliable.

As above with caregiving generally, research into dementia care appears to

undervalue social aspects. Care dyads reflect the social status of the care recipient as well as

the caregiver, but studies have only interrogated the influence of their biomedical condition

upon the caregiving experience. Beard and Fox suggest that social research encourages

“social disenfranchisement” of persons with dementia, and instead suggest a postmodern

31
outlook wherein the dementia is only a disability of the person’s relationship to memory or

time (2008:1510); this approach has tremendous implications for dementia caregivers as well,

and turns questions of wellness away from biomedical fixes and toward the broader social

support the care dyad receive.

The Dementia Care Dyad as a Social System

According to Bruhn and Rebach, “Caregiving is a process linked to the progress of

the care recipient’s decline” (2014:10), with stress and stress variance that are “interactional”

(156). Dementia care dyads exist in a mutually constrained tension where the actions and

impacts of one affect the other, perhaps more than any other form of caregiving. At the

heart of the care dyad is a constant question of prevention versus response, balancing

between the interests of a thinking, feeling adult — who may not realize or accept they are

sick and getting sicker — and the caregiver (and support system) that expects to outlive

them. Social shifts can occur to both the caregiver and the care recipient at the same time;

the less a person with dementia is capable of performing socially, the more responsibility

that falls upon the caregiver to either compensate or retreat. In a meta-ethnographic review

of international empirical studies and policy documents, U.K. researchers place a person

with dementia at the center of a large ecosystem where interactions with the caregiver, health

workers, and every level of society interact to affect their sense of personhood (Bosco et al.

2019). Similarly, Beard et al. endorse revisiting dementia as a "social process" to shift

dementia toward an "experience of memory loss rather than giving prominence to biological

and neuropsychiatric makers of decline" (2009:227). On the social significance of dementia,

an Irish study emphasizes, “Living with dementia cannot be isolated from interactions with

32
others,” (Greenwood et al. 2018:44, citing Daly et al. 2012). While the contexts of these

studies is beyond the scope of this thesis, their systematic viewpoints reiterate the social

significance not only of dementia, but of its care and of the greater support system of its

caregiver(s). Dementia care dyads already stand out in the literature; dementia care

demonstrates enough qualities of a distinct social system that many care studies separate

samples into two categories: dementia care and all others (see Pinquart & Sörensen 2003 for

examples). This social system approach warrants further exploration.

Perhaps it is not always the case that the care receiver’s age, gender, etc., influence

caregiving, but it is certainly true in dementia care, yet these social statuses are only

superficial facets of a dementia care system. Without a systemic framework, we cannot, for

example, uncover how the layers of socioeconomics within a care network may interact with

the socioeconomic trajectory of the person with dementia; dementia care may even be

ideologically sensitive: some families see dementia as either a mental illness (Stites et al. 2018)

or an inevitability of aging, even though neither are true.

At the heart of this social system is not only the care recipient, but the care dyad, a

“socially constructed dialectic” (Seaman 2018:64), wherein the caregiver and care recipient

must navigate one another’s realities and together navigate the “real” of others (Beard et al.

2009; Bruhn & Rebach 2014:154; Hermanns & Mastel-Smith 2012). Between its fuzzy

beginning (Seltzer & Li 1996) and its unavoidable end (Beard & Fox 2008), a dementia care

dyad builds, sustains, and ultimately loses an inner world of their own (Hermanns & Mastel-

Smith 2012; Seaman 2018). Not merely an alliance or social network (Cornwell & Schafer

2015), the depth and duration of this social entanglement fosters social consequences for

33
both members of the dyad and their external relationships as well (Schulz & Eden 2016:104-

106). Although linear and cyclical models of dementia care exist (Bruhn & Rebach

2014:157), the trajectory of dementia symptoms is usually erratic. Caregivers may lack

certainty from week to week, hour to hour, about when and how their loved one's cognitive

losses will manifest (Dunham & Cannon 2008), so adaptability is critical (Bruhn & Rebach

2014:163; Roth et al. 2015). Many caregivers report worrying about letting their guard down

at the wrong time (Seaman 2018), which can lead to hypervigilance, adrenal fatigue, and

burnout.

Social Linkages of the Dementia Care Dyad

The responsibilities of dementia caregivers extend far beyond mere ADLs and

IADLs (Pinquart & Sörensen 2004; Schulz & Eden 2016:76-77). Dementia caregivers

navigate ever-shifting dynamics (Berry et al. 2015; Bruhn & Rebach 2014:163) in a way that

tries to honor the present capabilities of the care recipient without exposing them to danger

or emotional harm (Beard et al. 2009; Berry et al. 2015). Serving as the care recipient's

primary (and sometimes only) buffer from the world, not only do caregivers decide whether

or how to intervene, they also determine when intervention must escalate and processes

must change (Dunham & Cannon 2008). A person with dementia may choose to pull back

from social engagement (Gauthier et al. 2013), but it is more common for their social role to

be withdrawn by others (Beard et al. 2009), leaving the care dyad to adapt roles and identities

on their own. Bosco et al. highlight social inclusion within their model of dignity, adding that

while dignity as a concept can be very individualized, “people know when they are not being

treated as such” (2019).

34
 The caregiver is generally expected to intervene if the care recipient is

uncomfortable, unsafe, or unruly (Berry & Apesoa-Varano 2017), reflecting respectively the

subjective feelings of the care recipient, the security of the care recipient (and possibly

others), and the subjective social expectations of any other persons around. Under this

dynamic, it falls to informal caregivers to "discipline and control [...] the aging body" and to

researchers to reduce caregiver strain; family caregivers, particularly adult children, can see

this power dynamic as a "role reversal" (Dunham & Cannon 2008:46, 52), and often describe

feeling ambivalent about it (Seaman 2018). Notably, people with dementia are mistreated

more often than any other elders (Schulz & Eden 2016:107; van Durme et al. 2012); they

also frequently experience depression or anxiety, although studies have not yet discerned

what portion of this pattern is cognitive (i.e., a symptom of dementia) as opposed to

psychosocial (Beard et al. 2009).

The dementia care dyad is a power dynamic that can be uncomfortable for both

members, but the erratic nature of dementia means there is no “script” available to help

individual caregivers balance the particulars of safety, ability, and agency for themselves and

their care recipients (Beard et al. 2009; Berry & Apesoa-Varano 2017; Gauthier et al. 2013;

Seltzer & Li 1996). Berry et al. (2015) distinguish stages of caregiving by whether dyad

actions are collaborative or unilateral, which is itself distinct from a spectrum of cooperation

versus resistance.

The perspectives of people with dementia are often overlooked in research; over 80

percent of people with dementia live “in the community” — that is, not in any form of care

facility (Hunt et al 2017:32) — and 8 percent receive no assistance (Berry & Apesoa-Varano

35
2017; Gaugler et al. 2019:31), yet Dunham and Cannon warn that dementia care studies

“often focus on the powerlessness of the care recipient” (2008:46). Beard and Fox argue that

“loss of self” for persons with dementia is “a dominant trope in America, reflecting a

postmodern disorientation and skepticism regarding time-honored conceptions of the

coherence and rationality of time, space, and selfhood” (2008:1510). The person

experiencing cognitive loss may themself resist the words “Alzheimer’s” or “dementia” and

— fearing “non-personhood” — delay diagnosis, preferring instead to blame “memory

complaints” on aging (Beard et al. 2009; Gauthier et al. 2013). While it may be easy to frame

the complicated dynamics of dementia care as the caregiver's problem, Beard and Fox

remind us that "the construction of an Alzheimer's identity is bi-directional, involving a

pendular process of maintaining past social roles and incorporating new ones" (2008:1511).

Persons with dementia often retain a sense of their own social lives (Beard & Fox

2008); impeded more by stigma than by ability (Greenwood et al. 2018), they remain

sensitive to rudeness, being ignored, and becoming isolated (Dunham & Cannon 2008).

Negative reactions — not only from strangers, but even from friends and loved ones — can

include talking down to the person, placing premature restrictions, or isolating them

(Gauthier et al. 2013; Johnson et al. 2015; Stites et al. 2018). Elder social ties are unique,

molded more by grief and health factors than employment or geography (Cornwell &

Schafer 2015), and especially vulnerable to race, gender, and socioeconomic status; persons

who face more and greater adversities have fewer and weaker connections than persons who

do not (Cornwell & Schafer 2015). Although persons with dementia who have strong social

support do not show slower cognitive decline than those with weaker support (Nichols et al.

36
2011), they may be less vulnerable to depression, anxiety, and other secondary consequences.

Feelings of guilt, denial, or loneliness can lead a person with dementia to overcompensate —

to “do what they can” (Berry et al. 2015) — however the person with dementia may have

very little power to affect the burden taken on by their caregiver.

For dyads who have active families or support systems, a caregiver must not only

make difficult decisions but justify them to others (Berry & Apesoa-Varano 2017; Seaman

2018); the primary caregiver is entrusted with learning the most about dementia and

disentangling the person from the disease for others (Beard & Fox 2008). Tensions arise not

only over what the caregiver can do but when, as many actions are warranted at one stage

but not others (Berry et al. 2015). Beard and Fox (2008:1510) invoke Kitwood and Sabat to

frame the unreasonable expectations held by persons without dementia as a “malignant

social psychology” (also cited in Dunham & Cannon 2008; Seaman 2018), ascribing moral

deficiency (cf. Mol 2002:7-13n) upon the person with dementia for having "excess

disability." Caregivers or others who attempt to bring the person with dementia to "reality"

are asserting a worldview that is confusing at best, placing their own needs above that of the

supposed care recipient (Beard et al. 2009; Werner et al. 2013).

Dunham and Cannon emphasize the "bio-power" of this dynamic, noting, “the

family caregiver is controlled by the same process of power/knowledge that disciplines the

aging body with dementia” (2008:46), while Seaman (2018:68) describes the dynamic as a

relationship of "non-action" that leaves caregivers feeling "tired, but not powerless." Very

little scholarship examines the assumptions and power dynamics at play, although this trend

is improving (Berry et al. 2015; Unson et al. 2015). Beard and Fox recommend,

37
“Accentuating the socio-moral necessity to care for people with dementia in ways that

recognize and preserve their dignity as human beings” (2008:1511). A care recipient's

resistance may have nothing to do with a caregiver's actual malfeasance, nonetheless it may

be their best method for disrupting power (Dunham & Cannon 2008). A long-term, primary

caregiver could be rejected at some random point, but in most cases they take on a singular

prominence (Dunham & Cannon 2008); by the late stages of dementia, this bond culminates

into a connection that is more instinctual than cognitive: when all others are forgotten, even

the self, this (nameless, story-less, just familiar) entity makes things better, solves problems,

and brings comfort.

Also overlooked in research is the extent to which the best interest of the care

recipient may conflict with the best interest of the caregiver (Hermanns & Mastel-Smith

2012). Dementia caregivers may experience not only isolation from being “stuck at home” or

having obligations with their time (Bruhn & Rebach 2014:109; Schulz & Eden 2016:103-

106), but a functional withdrawal of family and friends (Adler & Mehta 2014; Beard et al.

2009; Greenwood et al. 2018) who are unsure how to connect with the caregiver and/or be

around the care recipient. As loved ones pull away from the person with dementia,

caregivers, too, are subject to severe social isolation (Dunham & Cannon 2008; Seaman

2018), which may worsen if caregivers do not find solidarity with other caregivers (Adler &

Mehta 2014); what does that say about the caregiver if the care recipient becomes isolated

under the unspoken assumption that the person is already “lost” — is the caregiver lost as

well, or their efforts “inconsequential” (Beard & Fox 2008; Seaman 2018)? Prioritizing the

comfort of others may endanger the comfort of the person with dementia; conversely,

38
caregivers may risk alienating limited social support if they ignore the discomforts of

outsiders. Choosing between the two may exacerbate decision fatigue. To keep the person

with dementia safe and comfortable, caregivers may employ misdirection or even outright

lying (Berry et al. 2015; Dunham & Cannon 2008). Whatever challenges of selfhood or life

course the caregiver perceives from taking time out to give care, their actual sense of a social

role or sense of self is subject to being altered in the experience and its fluctuations (Beard &

Fox 2008; Bruhn & Rebach 2014:158-160, 163).

Dementia caregivers who report the greatest physical impact are not those whose

care recipients have the worst health, but those who began care with low social support

(Bruhn & Rebach 2014:162), yet dementia caregivers report difficulties receiving adequate

support (Hunt et al. 2017:3). Studies of elder social networks find that the greatest benefit

comes from those who can offer material support, though wellness monitoring through

regular interactions can also be beneficial (Cornwell & Schafer 2015). Families, employment,

and professional support (up to and including institutionalization) all involve systemic

interdependence that shifts along the care recipient's trajectory, but only professional

perspectives have received systemic analyses (Schulz & Eden 2016:106).

Though limited, findings suggest that informal caregiving can offer critical support,

foment family conflict, or both (Bruhn & Rebach 2014:155-160; Schulz & Eden 2016:106).

Dementia caregivers may relieve pressure upon a family by taking on care and

simultaneously failing to meet family expectations about their own employment, financial

success, or life goals (Eifert et al. 2016; Halberstam 2005:4-10). The paradigm of family care

is affected by shifts in the traditional family structure; few studies have explored the role of

39
divorce, step-children, and blended families in caregiving at all (Davey & Szinovacz 2008;

Schulz & Eden 2016:63), let alone in regards to memory-salient dementia care. Employment

has an effect on caregiving and caregiving has an effect on employment, though only

negative impacts seem to receive much attention (Schulz & Eden 2016:106). There may be

questions about the legitimacy of caregiving as a “career” for caregivers who leave other

jobs, or suspicion over shared finances (Bruhn & Rebach 2014:153-157; Hunt et al. 2015:62).

Social factors and social networks of the caregiver become intertwined, even

indistinguishable, from that of the care recipient, until only the caregiver’s merged support

network survives the death of the care recipient; however, not only their social support

systems interrelate.

The relevant practitioners who interact with the care dyad likely shift according to

the stages of the disease, although the literature does not address this. Over the course of

several years, the care recipient’s general practitioner may give over to a specialist, informal

support gives way to formal interventions from community or public programs, and paid in-

home support becomes institutionalization. Each of these transitions and forms of support

has been reviewed in literature by the relevant practitioner journals (Cornwell & Schafer

2015; Dam et al. 2016; Henning-Smith, Lahr, & Casey 2019; Parker et al. 2002), but they

have not yet been aggregated to center the dyad’s perspective. At heart of many of these

practitioner viewpoints is “biomedicalization” (Beard et al. 2009; Berry & Apesoa-Varano

2017; Spector et al. 2012; cf. Cockerham 2013; Mol 2002:7-13n), “an ever-complicated

elaboration and dispersal of biomedical infrastructures, innovations, and discourses”

(Seaman 2018:64). Foucaultian analyses of caregiver studies (such as Dunham & Cannon

40
2008; Roth et al. 2015; Seaman 2018) caution that many caregiver studies are merely an

extension of clinical dementia research and may serve interests of the clinic (Foucault

[1963]1994:117-122) without informing the interests of the populations being studied. The

caregiver must serve as a conduit between the relevant system and the care recipient; this

sort of case management is believed to be getting more difficult for all caregivers (Eifert et

al. 2016), and for dementia caregivers may be difficult to prioritize. That said, "person-

centered care" is gaining traction among intervention and medical practitioners (Whitlatch &

Orsulic-Jeras 2018:S59), which may help them meet caregivers partway; Schulz and Eden

(2016:220-223) endorse a model of "family-centered care" as defined by the National Quality

Forum, which includes "shared decision-making," convenience, and communication as

important facets.

In order to earn a place in research, informal care can no longer be regarded as an

isolated topic, but one belonging to a greater social context: its power dynamics, its history in

the nuclear family and biomedicine, and its epistemological quandaries do not belie but

exemplify the sort of interdisciplinary topics at which social research excels. To understand

informal caregivers in the U.S. is to come one step closer to understanding the people of the

U.S.

Conclusion

Dementia care is intensive and widespread, demonstrating complicated interactions

with the gender, race, marital status, age, and income of caregivers. Support that benefits the

caregiver will also benefit the care recipient; the inverse is true as well. Demographic and

socioeconomic factors for both members of the dyad are known to affect dyad health and

41
outcomes, but their cumulative interactions are rarely explored. Peer-reviewed articles on

caregiving and caregivers commonly emphasize interventions and practitioners, lacking not

only a longitudinal depth but also a sociological one and reinforcing simplistic views of

dementia care. Social and systemic perspectives are overlooked in care literature, particularly

for high-engagement families and atypical caregivers, and even analyses of interdependence

and power have yet to solidify the importance and relevance of caregiving to U.S. society.

The symptoms of dementia are experienced more socially than physically, yet

biomedicalization has sidelined social inquiry. Social factors of caregiving are likely

mechanisms of this system, but require deeper, more multidimensional research. The

opportunity exists to develop nuance in caregiving research by centering not the caregiver or

care recipient, but the mutual dyad, and to view that dyad as one system operating among

many toward greater Durkheimian solidarity (George & Ferraro 2016). With a systemic

outlook on dementia care dyads and their support systems, researchers can expand our

understanding of informal social systems, further distinguish dementia care from other

health and intimate dynamics, and close research gaps.

THEORETICAL FRAMEWORK

While trying to reconcile tension between white feminists entrenched in academia

and Black feminists who wanted "concrete action," bell hooks (1991:5) makes a case for

theorization. She cautions that this work could be easily hidden or co-opted, but that the

work is necessary to break down the "false dichotomy between theory and practice" in

feminism and support a throughline from individuals' childhood traumas to their "healing"

42
group actions in the present and investing toward collective progress in the future (hooks

1991:5).

In academia today, there is no dichotomy for caregiving literature because there is so

little theory being applied (Dunham & Cannon 2008). If the work of scholarship is to inform

pathological outcomes or financial impacts, there is no shortage of studies that will answer

these questions with statistical precision; however, if the work of scholarship is to recognize

something about the human experience in how and why caregivers exist — now, not one

hundred years ago nor a hundred years in the future — and if that care scholarship is at all to

be healing, it needs to reconcile theory with practice, and like hooks it necessitates

"engagement with collective sources" (1991:3) While a wide range of caregiver studies do

exist, their theoretical grounding is just as extreme as their sample sizes: so broad that no

depth can be gained or so narrow that no impact can be measured. For every study applying

broad social psychology to caregivers, there is a new and distinct theory targeting caregivers,

only to be forgotten after one or two studies. In between, there is nothing about how

individuals engage institutions or small groups in the caregiving context. It is feasible that

caregiving research could thrive if it had the structure and accountability of a theoretical

canon or a discipline overseeing its rigor; as it has neither, however, care scholarship

demonstrably suffers.

This review of theories makes the case that care research is a social phenomenon and

could be implanted in sociology without slowing from existing research trajectories. Multi-

layer frameworks require versatile theories, but with sociological designs scholars can expand

care research in every direction. As the power dynamics of dementia care specifically are

43
explored, the need for a critical perspective emerges, with the post-Foucault model of queer

subjectivity (Halberstam 2005) and the multi-contextual model of intersectionality

(Crenshaw 1991, 1993) selected for the present study.

Curing vs. Caring

Social inquiries must confront the question of positivism: whether truths uncovered

are fully knowable or probabilistic at best (Moradi & Grzanka 2017). Even so-called “hard”

sciences now make room for post-positivistic stances, but the divide itself is still quite binary;

one cannot study planets in the same way one studies quanta because the physics function

differently, so scholars must specify their scope. When social scientists design a study, that

study usually takes either a quantitative or a qualitative stance. These viewpoints are

specialized to extreme contexts of inquiry and only rarely and with great care can qualitative

and quantitative methods be applied to the same subject at the same time. As Adler and

Mehta assert, “in the real world things are ‘connected’” (2014:8). Outside of mixed methods,

which are rare and cumbersome, it is theory that helps connect the different stances and

show their relevance to one another. Sociology explicitly includes all levels of social

engagement and has set many precedents for linking different contexts to one another,

making it ideal for developing a systematic read of caregivers. We see that caregiving

research favors either the biomedical/demographic stance (quantitative) or the narrow,

situational stance (qualitative), but developing their relationship to one another (and to

greater society) requires theories and frameworks that can bridge the micro and the macro.

What is needed is a theoretical approach that can connect many layers at once from the

national demographics to the personal engagement of the care dyad.

44
 The relationship between caregiving and sociology need not be one-way. In making

their case for aging research, George and Ferraro summarize the social sciences as asking,

“what mechanisms convince societal members to create and sustain societies, even when

those mechanisms require members to sacrifice some of their own resources, gains, and

autonomy?” (2015:np). This framing perfectly describes the impact caregiving research can

offer sociology, and may even lead to new, non-bureaucratic answers to Durkheim's

fundamental sociological question of solidarity. George and Ferraro go on to state, “no

matter how much structural arrangements appear to differ across societies, the functions

they serve are the same” (2015:np). This invites the sort of multi-modal frameworks where

sociology works best.

The review above demonstrates that care research tends to polarize toward the

macro- and micro-levels, with little development in-between, nor momentum to fill the gaps

(George & Ferraro 2016). This pattern marginalizes caregiving as a phenomenon that is

studied as solely pathological — or worse, as a solely logistical response to pathology (Schulz

& Eden 2016:163). Apesoa-Varano, Barker, and Hinton refer to this divide as “curing”

versus “caring” (2011), and Seaman indicates a further discernment is occurring between

“caring for” and “caring about” (2018:65). Here, “curing” is broad shorthand for the

biomedical model, a conquering of bodies rooted in nineteenth-century bureaucracy

(Foucault [1963]1994:64-82; Lock 2002:33), which necessitates a certain approach to

categorizing, studying, and treating that which is errant. Theoretical reflections on how

biomedicine is enacted find that the emphasis remains binaristic — sick or not sick (Mol

2002:7-13n) — without looking at patients holistically or socially (Cockerham 2013). From

45
this logistical (curing, or curing-adjacent, since no cure exists for dementia) outlook, it may

make sense to honor life by prolonging it, increasing a prominent quantitative measure, and

therefore to assess caregivers by their contribution to prolongment and measure

interventions by their measurable improvements to caregivers' measured outcomes. In other

words, there is a lot of measuring and very little attention to care as a social or systemic

process. A logistical approach is direct and simple, finding facts on the ground as they are.

There is no need to challenge statistical conventions, nor look beyond the most

numerous/visible/normative populations for impact or significance — count the big groups

and delimit the "others." A logistical approach leaves no room, nor has any need, for

disciplinary tradition or theoretical framework, but neither does it offer any indication what

is to be measured when life cannot be prolonged, as is currently the case for persons with

dementia.

And yet, even a logistical viewpoint calls for better data than purely quantitative

studies have been able to provide. Caregiving scholarship often suffers from inconsistent

rigor and unresolved contradictions, leaving very few studies that can be compared to one

another directly (Davey & Szinovacz 2008; Pinquart & Sörensen 2003; Schulz & Eden

2016:45). Even Medical Sociology — rooted as the practitioner's ally (Cockerham & Ritchey

1997; Mol 2002:7-13n) — has advanced beyond mere logistics to recognize the need for

studies of patients' social factors, because disparities effect a measurable difference upon

wellness (Cockerham 2013).

Caregiving scholarship, on the other hand, often fails to challenge methodological or

even popular assumptions about caregiving. Researchers echo communities in viewing

46
caregiving as an isolated, biomedical, aging-related, and inherently negative phenomenon to

be fixed rather than a collective, social, multi-faceted, and mixed/positive phenomenon to

be understood and enhanced (George & Ferraro 2016; Roth et al. 2015). Studies often

measure outcomes for the care recipient or for the caregiver but not both (Schulz & Eden

2016:106). While macro-level research tends to compartmentalize caregivers into binary

boxes (formal/informal, primary/secondary, employed/not employed), the broader

implications of 16 million people a year dedicating a significant portion of their lives to

something other than income or procreation are rarely explored. With each caregiving study

serving a different research tradition, gaps and inequalities can always be someone else's

problem. This trajectory leaves not one rich and intricate canon to be organized and

developed, but several pools of oppositional chaos; here, studies of high and low quality

alike receive very little attention or follow-up because they have no way to stand out.

Discomfort on the part of would-be research participants is seen as a failing of the recruit,

not of the research itself (Bruhn & Rebach 2014:110). Further, biomedical inquiries might

discern that social support helps improve outcomes for the care dyad, but examining the

qualities of that support falls outside their research scope.

Both logistical and social considerations matter in caregiving. Unfortunately, the

tension between “medical” and “social” priorities is not even new in caregiving: a 1980

article from the Rand Corporation warns that discussions over long-term care carry the same

tension, and that favoring the biomedical perspective alone creates a setback for social policy

(Kane & Kane 1980). Invoking “heroic medical measures” and disability, the authors specify,

“The very permanence and intractability of these problems argue for societal provisions to

47
protect the elderly from a permanent patient role for decades before their death” (Kane &

Kane 1980:241), and yet this very phenomenon is a growing reality for U.S. elders (Eifert et

al. 2016). Forty years later and still without a firm methodology or design structure,

caregiving research remains disconnected from social and economic topics, even from its

own context — and it shows little sign of dramatically improving the lives of caregivers.

Many social inequalities continue or are even exacerbated in a caregiving context. Caregiving

continues, research continues, but they do not connect.

Caregiving research has some footing in several disciplines — of which nursing, itself

sometimes seen as an elevated version of caregiving, may be the best fit (Bruhn & Rebach

2014:22) — and even within social theory, but efforts to unify all facets of caregiving are

rare. In 2014, Bruhn and Rebach published The Sociology of Caregiving, a textbook attempting

to pull together many of the disparate research threads. While it is a useful compilation of

research findings up to that point, it suffers many of the logistical limitations discussed

above: efforts to place caregiving in a global context rely on numbers and scale, with

significantly less attention to the lived experiences of caregivers or the interconnectedness of

layers. The authors rightly point out how social inequalities interact with care dynamics

(2014:79-119), but only in the process of attempting to organize the chaos — not to apply it

forward and further understanding. Continuing the pitfall of serving pathological interests

through simplistic designs, many facts and figures are presented without commentary. No

efforts are made to reconcile disparate methods or address academic distribution of care

topics (in fact, while there are chapters devoted to policy and social media, no chapter or

segment reviews caregiving research methodology). Perhaps the weakness of this text is a

48
continuation of the sidelining of caregiving as only a tangent to social problems, not a social

problem or social system in its own right, but more likely it is simply trying to do too much

with a subject that lacks cohesion. Caregivers are not a monolith (Cook et al. 2018;

Montgomery & Kosloski 2013; Morgan et al. 2011), and treating all care dynamics the same

is perhaps a better exercise for research on altruism or anthropology (Roth et al. 2015;

Seaman 2018) than for sociology. To see the systematic-social, it is necessary to recognize

that disease itself can be social (Beard et al. 2009), that dyads interact with and outside of

families and larger support networks (Bosco et al. 2019), and that caregiving looks

dramatically different for a broken leg than it does for dementia. Before a holistic viewpoint

can be developed, perhaps fitting care experiences into a spectrum or small matrix, some of

the most distinct care dynamics must be outlined more clearly (Brodaty & Donkin 2009).

This study takes the first step by separating dementia care and deepening its quantitative and

literary context.

To keep caregiving on the periphery of biomedical research is to overlook the

inherent discord of priorities; the goal of medicine — to prolong life — is secondary at best

to the dementia caregiver, while quality of life is secondary at best to models of pathological

treatment (Foucault [1963]1994:117-122). Previous attempts to theorize about caregiving

have come from the life course perspective, feminist ethics of care, and social role theory

(though not other approaches from Sociological Social Psychology), but none has succeeded

in attaching caregiving as an important component of theorization, nor as a niche worthy of

ongoing study. Life course perspectives tend to reiterate that, aside from parenting, some

level of caregiving is associated with old age (Bruhn & Rebach 2014:41-42), but this

49
overlooks demographic shifts and cultural contexts — again, caregiving as we know it exists

in this time and place for a reason. Scholars of ethics of care work to elevate the study of

compassion and qualitative judgment to an academic level, but struggle to disentangle the

construct from gender essentialist implications and sharpen the distinction between literal

and metaphorical “care” (Barnes & Brannelly 2008; Jordan 2020). Role theorists look to

discrepancies (which affect measures of caregiver burden) rather than consistencies (Bruhn

& Rebach 2014:8-9, 157; Savundranayagum & Montgomery 2010). Other theories that have

gained traction include predicting caregiver outcomes with the stress-process model (Moon

& Dilworth-Anderson 2015; Nichols et al. 2011) and development of a caregiver identity

theory (Friedemann & Buckwalter 2014; Montgomery & Kosloski 2013), but they are

designed once again to support interventions, to meet curing (or curing-adjacent) priorities.

These findings tend to speak to the role of the caregiver within a family or a biomedical

context, but not within a community or society at large.

To break away from the curing vs. caring tension requires a theoretical framework

that can be applied across the macro/micro disconnect, that is just as concerned with large-

scale disparities as it is with individual outcomes, and ideally that also considers the agency

and quality of life of the person with dementia (Berry et al. 2015). Having rejected the

concept of victory over a disease that cannot yet be stopped, this study approaches

Foucaultian/post-structural priorities (i.e., that research concerns cannot be further

addressed without decontextualizing and recontextualizing the social powers that led to this

point). While post-structuralism is not inherently anti-intervention, it does center alternative

50
questions about why and how current intervention models even exist (i.e., in whose

service?). The only theories that meet all these criteria are critical theories.

Critical Theories: Queer Subjectivity and Intersectionality

By applying a critical approach, we allow for the possibility that caregivers’ social

experiences may run deeper than a mere lack of social support. It is not enough that

caregivers are doing too much work or that they feel isolated doing so; they are being

systematically excluded from societal relevance by a society that has no place for them and

fails to contextualize their endeavors beyond economic measures. Though individual

caregivers may be regarded highly for their contributions, caregiving is not a common nor

comfortable topic in the U.S. because it is incongruous with expectations and values. Fraser

invokes capitalistic roots in a growing “crisis of care” and how the familial care gap is being

filled for the Global North with immigrants from the Global South (2016:112; see also

Glenn 2010:181; Halberstam 2005:19; Scales et al. 2020). Not only are existing inequalities

being exacerbated by the work of caregiving, but new ones are being created.

Seaman also points out that in the biomedicalization model, the future has become

"the site where solutions to our medical ills lie," and cautions that all dementia care work

could be invalidated if/when a cure is found (2018). The sense of time as experienced by

caregivers is a good example of just how far removed from “everyday” society many

caregivers are. In a series of 2011 blog posts for National Public Radio, astrophysicist Adam

Frank approaches social science in discussing how one’s relationship with time is “particular

to our culture and our history” (2011a). Frank blames contemporary “time-logic” on

industrial concepts of efficiency, linking it to contemporary challenges as climate change,

51
over-consumption, and precarious economies (2011c). Frank further postulates that, “As

individuals, the desire to build a new [time-logic] springs from our deeply felt need to reclaim

value and balance in our lives” (2011a:np, emphasis in original). Essentially, what Frank asserts

is that time is social, and in doing so, he implies that anyone who does not participate in the

(economic, work-centered, mainstream) time — what he calls a “time-bender” (2011b) — is

anti-social.

Queer theories also address anti-social relationships with time. Halberstam suggests

that alongside the centering of work-time, there is also a “reproductive temporality” and that

deviation — any deviation — constitutes “queer time” (2005:4-6; cf. George & Ferraro

2016). Because there is no room in the normative society for anyone who “leaves the

temporal frames of bourgeois reproduction and family, longevity, risk/safety, and

inheritance,” such a person is othered; temporally, they are “queer” irrespective of their

sexual or gender identity (Halberstam 2005:6; Halberstam is here deploying “queer” as a

label of vague/unsortable and transgressive outsider-dom). Queer time might especially fit

someone who exits a traditional (employed, procreative) life course to take care of someone

with neither reproductive nor income potential; unlike marginalized persons or directly

experienced hardships, a caregiver has some level of choice in the matter (even if the choice

is between becoming a caregiver or letting a loved one decline alone and unsupervised).

It turns out there is a lot separating caregivers from non-caregivers. Writing

specifically about dementia care, Beard et al. remind us that efforts to “fix” persons with

dementia are “predicated on the superiority of present-time orientation, where the goal is to

restore ‘reality’” (2009:228). The authors discourage “enforcing our cognitive reality by

52
letting go of normative expectations and ‘joining’ people with dementia in their worlds”

(Beard et al. 2009:228), and in doing so suggest there may be nothing wrong with not fitting

the common time-logic at all. To be clear, most caregivers struggle with their care recipients’

altered reality and probably would retain a role in the dominant time-logic, but it is not the

person with dementia who isolates them for the deviation; the assertion of a single and

universal “way” of being in the world means that those existing in queer time — whether or

not by choice — must be marginalized, lest the practice spread further (Halberstam 2005:10-

12, cf. Montgomery & Kosloski 2013; Schulz & Eden 2016:105). This is how you get queer

subjectivity.

Halberstam also introduces “queer time” as another facet of queer subjectivity,

describing queer space as “new understandings of space enabled by the production of queer

counterpublics” (2005:6). Queer spaces may be rooted in non-heteronormative sexuality, but

what is created there is not just for LGBTQ people; it certainly is not for everyone (there is

still a normative, cisheterocapitalist reproduction cycle out there, humming along, from

which queer subjects are discarded), but it might be for anyone. Queer subjectivity, then, is

queering the definition of “queer” itself (Hall 2003).

Recall that dementia caregivers live as much or more in the world of their care

recipient as their own; their understandings of space and time become fluid, not entirely

their own, and that others’ positivistic constructs of reality are irrelevant at best (harmful at

worst). Whether or not an individual caregiver applies some definition of “queerness” to

their life, there is room for their counterpublic in queer space and time, themselves forged

and scarred by the AIDS crisis, its victims, and its caregivers (Halberstam 2005:2); witnesses

53
to dementia in particular are familiar with a “constantly diminishing future.” Queer

subjectivity may apply to any caregiving situation, but especially to those who delay or

sacrifice normative life courses that were otherwise available. As a research framework

(which Halberstam uses elsewhere in the same book to develop counternormative archives

for Brandon Teena and other queer narratives that had been altered and accepted in

normative media), it says first and foremost that there is valuable information in non-

normative experiences; these experiences are not always a matter of failings or of limited

opportunities (although caregivers and their prior employment prospects comprise a

research gap waiting to be filled!), but of choices that are not recognized. Queer subjectivity

is taking the opportunity to give illicit recognition, to see the unseen, and perhaps to allow

them to tell their own stories (Seaman 2018).

Because queer subjectivity is posited as more of an interpretive stance than a

research orientation, a clear methodology is vital to its application. Scholars who call for

more “specialized categories of caregivers” (Brodaty & Donkin 2009:224) specifically link

this need to growing evidence of health inequality (Cook et al. 2018). Both quantitative and

qualitative approaches risk repeating societal erasure and diminish academic rigor when they

look at gender, race, and class as separate phenomena from health status, quality of life, or

caregiving outcomes. Developing such categories would require studies to both identify

significant patterns across large surveys of caregivers and offer context for the patterns as

measured. Public datasets allow and invite secondary inquiry to clarify these patterns, but

giving them context will require a theoretical foundation that is both sensitive to the variety

of care dyads and recognizes their origins in broader hegemony. Fortunately, studies are

54
calling for more intersectional perspectives in health and aging research (Evans 2019; Ferrer

et al. 2017; Gkiouleka et al. 2018; Hankivsky 2012; Richardson & Brown 2016).

Intersectionality examines the complicated, often material, outcomes for persons

who face multiple forms of systemic oppression at the same time (Crenshaw 1991; Hancock

2016:12). In the simplest terms, persons who face both racism and sexism in the workplace

demonstrate lower wages and fewer promotions than those who face racism or sexism alone

(Crenshaw 1991), but these forces also have implications upon activism, politics, and social

science (Crenshaw 1993; Hancock 2016:7-11; Moradi & Grzanka 2017). Intersectionality

asserts that not only is it significant to recognize when and how these disparities amplify one

another, but that systems of redress refusing to acknowledge more than one form of

discrimination at a time are themselves sources of systemic oppression (Crenshaw 1991,

1993). Intersectionality offers an inversion of social stratification, the positivistic sociological

theories collectively asserting that we can come to understand macro-social power dynamics,

if only through ever-more precise (and complex) matrixes of separation (Grusky &

Weisshaar 2014:4-15, 45, see also Tumin 37 in the same volume; Wright 2005:12-21, 26, see

also Breen 41 in the same volume); instead, intersectionality asserts that to comprehend the

magnitude of a person’s “total” marginalization, you do not divide the forces against them

and add them up, but combine them and pay attention to their cumulative, qualitative impact

(Crenshaw 1991, 1993; Moradi & Grzanka 2017).

Intersectionality is the culmination of a hundred years of Black feminist thought,

seeded by the time Anna Julia Cooper wrote in 1892 of the “unique position” of Black

women as not merely Black and not merely women, but “confronted by both a woman

55
question and a race problem” (Guy-Sheftall 1995:45; cf. Crenshaw 1991). The Combahee

River Collective expanded this sentiment in 1977, stating, “We [...] find it difficult to separate

race from class from sex oppression because in our lives they are most often experienced

simultaneously” (Guy-Sheftall 1995:234; cf. Moradi & Grzanka 2017). Kimberlé Williams

Crenshaw named intersectionality and cemented its academic stature in two law articles in

the early 1990s, situating the need for multifaceted frameworks against oversimplified legal

battles over equal pay and censorship (1991, 1993). While its applications comprise any

situation where social marginalizations do not merely accumulate but create unique and

complicated disparities, intersectionality is founded and defended through the tradition of

Black feminists of the U.S.

The face value for quantitative applications may seem low because intersectionality

has developed as an interpretive (Hancock 2016:12-13) practice of critical resistance against

discrimination, harmful binaries, and having to compromise to pursue recourse and identity.

Bringing this interpretive framework is not a simple matter of mathematically adding

together several disparities and calculating their cumulative harm, and theorists fearing

dilution and appropriation now push back against scholars who conflate intersectionality

with any study invoking multi-layered identities (Moradi & Grzanka 2017). What matters is

not whether there is a mathematical approach, but where the math is rooted. Material

differences do exist, they can be traced to social and systemic discrimination, and they can be

calculated for a specific population, but this process must originate from the population

itself (Moradi & Grzanka 2017). Intersectionality does not resist quantitative inquiry, only

the quantitative predilection for rigid standardization. Oppressive forces are in a constant

56
state of flux (indeed, perhaps time — or Queer Time — should be added to the

intersectionality toolbox, but this question is outside the scope of the present study), so no

scholar can ever say they have identified the textbook formula for intersectional enumeration

for all caregivers; we can only say that we have observed certain patterns within a particular

population at a particular time and, from there, calculate the (socio-)economic impact.

The call for intersectional analyses to hone and apply quantitative methods is

growing (Bauer & Scheim 2019), including in health contexts (Green, Evans, & Subramanian

2017; Hankivsky 2012; Richman & Zucker 2019; Warner & Brown 2011). Combining

intersectionality and statistical analysis will not only improve the quality of statistical

methods (by, for example, contextualizing systems of oppression in interaction effects), but

also continue the advance of feminist inquiry beyond theory and ontology (Scott & Siltanen

2017). Importantly, as argued by Ange-Marie Hancock in her Black-woman-led review of

intersectional topics, “intersectionality-like thinking about how power is relationally

constituted predates and anticipates Michel Foucault’s well-known arguments about power”

(2016:164; cf. Foucault [1963]1994:99-100). Studies that look at, for example, race and

gender together can improve the present but also elevate the history (individual and

collective) of participants (Richardson & Brown 2016), which will also expand other critical

standpoints.

Applying intersectional approaches to caregivers is not only helpful but necessary,

because “unique characteristics of various racial and ethnic minorities interact with the

challenges common to all [care situations] and result in substantially different outcomes”

(Angel et al. 2015). When scholars treat caregiving as a gendered phenomenon, they obscure

57
the ways multiple marginalizations place the disadvantaged caring for the advantaged; in the

middle class, it may be women caring for men, but in long-term care facilities it is the

working class caring for the middle class; in cities it is the less educated caring for the highly

educated; in many states, it is people of color caring for whites; and across the country it is

often the immigrant caring for the citizen (Glenn 2010:181; Scales et al. 2020). These

oppressive dynamics do not operate separately, but in constant interaction with one another,

rooted in many of the same assumptions about a positivistic, right or wrong world; this is

why intersectionality must remain interpretive and contextual, and why caregiving must be

seen as more than just a gendered phenomenon.

Conclusion

This section has reviewed some of the theoretical frameworks that have previously

been applied to dementia care and caregiving research at large and detailed the challenges

that have emerged. In promoting a sociological research stance toward dementia caregiving,

this study moves away from interventionist perspectives that have little interest in the

support systems of caregivers and has faltered in the face of health inequalities. Critical

theories are suggested as an alternative that can meet the needs of dementia caregivers, their

care recipients, and practitioners of all levels. Halberstam’s queer subjectivity and

intersectionality are selected to advance the current analysis through inquiry of how othering

may occur against and between informal dementia caregivers.

HYPOTHESES

This study examines the effect of intersecting social marginalizations upon dementia

caregivers, themselves a distinct (potentially othered) subset among informal caregivers. An

58
existing quantitative dataset is utilized to demonstrate that critical theories can be applied to

data collected in a straightforward, positivistic manner. The following three hypotheses will

test whether social disparities demonstrate a significant impact upon dementia caregivers; to

a lesser extent, they will also explore the potential for sub-categorizing dementia caregivers

according to advantages and disadvantages in their care circumstances.

H1: The age of dementia onset among care recipients varies by race. This hypothesis places a

pathological benchmark in the context of social marginalization. Generally, dementia is

considered “early onset” if it is diagnosed before age 65, but a racial disparity that favors

whites could indicate that people of color are being ignored or systematically excluded in the

development of pathological benchmarks. This hypothesis will be tested through cross-

tabulation and chi-squared and its relevance explored.

H2: Dementia caregivers with normative characteristics will demonstrate less physical strain than

non-normative caregivers. This hypothesis tests whether reports of low physical strain are

significantly concentrated in caregivers whose demographics are most common or average.

Physical strain tends to be one of the lower burden measures among dementia caregivers;

however, it is not often stratified by social factors. Factored in a multivariate regression

model, these social indicators may demonstrate that normative caregivers have an outsized

influence on means, obscuring the physical strain reported by non-normative caregivers, or

those othered for having few of the following normative traits: white, woman, over 50,

higher income, not employed, caring for a parent.

H3: Dementia caregivers with higher social marginalization will demonstrate a higher ADL index

than other caregivers. This hypothesis tests whether the ADL index measures significantly

59
higher among caregivers who face more intersections of social adversity — in other words, a

measure of care intensity will be assessed to determine whether caregivers who are othered

by society are othered amongst caregivers. ADLs are often counted as an “objective”

measure of care burden, but this hypothesis asserts that care dyads who face historical and

ongoing disadvantages will see health and care disparities reflected in their care challenges.

Multivariate linear regression will be employed to identify a significant concentration of

socially marginalized caregivers and compare their ADL indexes against caregivers with the

fewest marginalized traits: person of color, woman, lower income, less education, employed,

solo, and sandwich generation.

60
 CHAPTER III

DATA AND METHODS

This chapter introduces the secondary dataset to be employed and a description of

care variables utilized. Statistical methods (correlation matrixes and multivariate regression)

are discussed and limitations of the study are discussed.

DATA

This study utilizes Caregiving in the U.S., a periodic, public interest survey of adult

caregivers that includes both objective and subjective items about the care dyad. Four waves

of this dataset exist (1997, 2003, 2009, and 2015, although 2015 employs a new

methodology) and are available to the public for download without registration. A new wave

is being released in 2020. Data related to dyad demographics, characteristics, self-reported

stress, and an index of total ADLs were examined within the 2015 wave (conducted in late

2014) via SPSS 25.

Sample

From a random online sample of 7,660 nationwide adults, 1,248 online surveys were

completed by adults who had given care to some other adult in the prior 12 months

(NAC/AARP 2015). Oversamples are not included as they fall outside the scope of this

study. The sample is weighted according to the specifications recommended by the dataset’s

methodology and codebook, resulting in a dementia caregiver population of N = 259.

61
VARIABLES

Cases have been restricted to dementia caregivers who answered all necessary items

for the dependent variables. This section describes the variables as originally coded and

initial modifications made therefrom; invalid responses (Not Sure/Don’t Know or Refused) are

omitted.

Dependent Variables

For dependent variables, this study utilizes one indicator of dementia scope and

three indicators (one objective and two subjective) of care burden. For the scope of

dementia, age of onset estimates when the care recipient began to receive care for dementia;

this interval-ratio variable, measured in years, was calculated by subtracting the survey’s item

for duration of care from the age of the care recipient. This calculation does not recognize

potential changes in caregivers or situations where the care recipient previously received care

for something other than dementia, yet it is the closest we can get with the available data to

age of diagnosis or visible onset; a dichotomous version tracks those who are estimated to

have experienced early onset (Age 65 or Over = 0, Age 18-64 = 1).

For subjective caregiver burden, two survey questions record physical strain or

emotional stress from caregiving as reported by caregivers. For both 5-level scales, the

lowest item (recorded as 1) was Not a strain at all (Physical Strain) or Not at all stressful

(Emotional Stress) and the highest item (recorded as 5) was very much a strain or very stressful;

the available responses approximate respondent attitudes but cannot be quantified rationally,

so they are ordinal variables.

62
 The remaining dependent variable is an index of ADLs as counted by NAC/AARP,

based on cumulative (recorded as No = 1 and Yes = 1, but here modified so that No = 0, Yes

= 1) responses to the following items:

● Getting in and out of beds and chairs

● Getting dressed
● Getting to and from the toilet
● Bathing or showering
● By dealing with incontinence or diapers
● By feeding him or her

This dataset records a total of six ADLs, so the range for ADL count is 0-6; the higher the

count, the more types of daily support the caregiver delivers. Since the severity of ADLs are

not quantifiable as regular intervals, this index constitutes an ordinal variable.

Note here that other studies tend to combine the mobility measures — getting in and

out of beds and chairs and getting to and from the toilet — into one item, resulting in only five total

ADLs; NAC and AARP do not explain this methodological decision, but do report a 16-

point difference between the mobility measures as separated (Hunt et al. 2015:40).

Table 1. Dependent Variables Used in Analysis of Social Aspects of

Dementia Care Dyads, 2015 (NAC/AARP)
Variable Mean Std. Dev. Skew t
Self-Reported Physical Strain (5-Level) 1.7364 1.26160 .218 22.161
Self-Reported Emotional Strain (5-Level) 2.4525 1.17694 -.305 33.553
ADL Count (0-6) 2.1470 2.13535 .573 16.189
Age of Onset (Years) 72.3724 16.58402 -1.583 70.267

Independent Variables

The following dyad characteristics serve as independent variables: relationship,

caregiver age, caregiver primary status, and caregiver employment status. Relationship is

recorded as a 36-item categorical variable referring to the relationship between the dyad —

63
whether they are partners (coded as 4: Companion/Partner, 20: Spouse, or 36: Same-Sex Partner),

parent/child (coded as 6: Father, 7: Father-in-Law, 13: Mother, or 14: Mother-in-Law), some

other type of relative, or not related (coded as 24: Friend, 25: Guardianee, 26: Neighbor, or 27:

Other Non-relative) — based on responses to the survey question “Who are you caring/did

you care for?” Three dyad relationships (Family, Partnered, and Parent) are recoded

dichotomously (No = 0, Yes = 1) as needed. Primary is a dichotomous variable initially

calculated ordinally (Sole Caregiver = 1, Primary but not sole = 2, Someone else is primary = 3, and

Shared equally = 4) that reports whether the respondent served as primary caregiver (No = 0,

Yes = 1); this study also created a dichotomous (No = 0, Yes = 1) variable for solo caregivers

(all solo caregivers are primaries, but not all primaries are solo), and inverted solo to report

whether a caregiver has Unpaid (informal) Support (No = 0, Yes = 1). Employment is a

dichotomous variable (initially recorded 1 = Employed, 2 = Not Employed, recoded No = 0, Yes

= 1) reporting whether the caregiver says they were employed and caregiving simultaneously

at any time during the previous year.

Intervening Variables

The following caregiver demographics serve as intervening variables: gender, race,

education attainment, household income, and caregiver age. As women are central to two of

the hypotheses, gender is here discussed as a dichotomous variable for woman (initially

coded as Male = 1, Female = 2, here recoded so that Man = 0 and Woman = 1; see

Introduction for notes on terms of inclusion). Race was recorded across multiple survey

questions and merged in data collection as a 5-item categorical variable (White = 1, Black = 2,

Asian = 3, Other = 4, and Hispanic = 5); for this analysis, two dichotomous variables report

64
when a caregiver is white (Person of Color = 0, White = 1) or a person of color (White = 0,

Person of Color = 1), and a third dichotomous variable reports on Black and non-Black people

of color (Non-Black POC = 0, Black = 1). Education attainment depicts highest education

attainment (initially coded ordinally as Less than High School = 1, High School grad/GED = 2,

Some College = 3, Technical School = 4, College grad = 5, Graduate school/Grad work = 7) and is

here recoded dichotomously based on whether the caregiver has attained some certification

beyond high school diploma (High School Diploma or Less = 0, Trade/College/Graduate-level

Certification = 1); as will be discussed in the next chapter, another dichotomous variable was

created for caregivers reporting the least education (High School Diploma or Beyond = 0, No

Diploma = 1). Household income (in dollars, initially merged from several questions into the

following near-ordinal categories: Under 15000 = 1, 15000-29999 = 2, 30000-49999 = 3,

50000-74999 = 4, 75000-99999 = 5, 100000 or more = 6; codes 9 and 10 report incomes that

were, respectively, less than or more than 50000 but otherwise unspecified) is presented here

dichotomously (Less than 50000 = 0, 50000 or More = 1); note that the U.S. national average

household income for 2014 was $53,657 (DeNavas-Walt & Proctor 2015). Caregiver age was

recorded in years as an interval-ratio variable, but is here presented dichotomously (Ages 50+

= 0, 18-49 = 1); an inversion of this variable (Ages 18-49 = 0, 50+ = 1) is also utilized for

older caregiver.

DATA ANALYSIS

Bivariate Correlations

Cross-tabulation is a specific type of correlation matrix specialized for unmatched

levels of measurement across only a few categories. Cross-tabulation measures the

65
significance — but not the strength — of superficial relationships by measuring correlation

between all categories within the variables compared. Because this method involves a

correlation matrix that subdivides into smaller batches, some variable categories have been

merged or omitted to fewer levels per statistical convention. H1 is tested utilizing a

controlled cross-tabulation.

Multivariate Linear Regressions

Multivariate linear regression is a statistical technique that determines whether a

dependent variable can be predicted by some combination of independent and intervening

variables. Statistically significant regression models report the strength and direction of a

predicted relationship, as well as its significance; errors between expected and actual

outcomes are minimized by squaring and adding them together and letting the lowest

possible total determine the effectiveness of the regression (this technique is called ordinary

least squares). These interactions are relevant to intersectional concepts because models of

co-occurring variables can demonstrate an impact upon a dependent variable that is more (or

more complicated) than the sum of their individual interactions (Evans 2019:95); such

models can be merged into a single variable (called an interaction effect) for further analysis.

By exploring data in this manner, complex patterns become predictions, explaining variance

within a specific population (such as subsets of dementia caregivers).

Multivariate linear regression measures the relationship between interval-ratio

variables. Dichotomous variables, if prepared appropriately, can also undergo regression

(with each other or with interval-ratio variables) as dummy variables. A dummy variable

must be stated strictly in terms of a single category (i.e., “dummy” and “not-dummy”), with

66
the largest category conventionally serving as referent (“not-dummy”). Regressions can

theoretically consider many variables at once, but when significance is equal, a simple

solution is preferred over an exhaustive one.

The two remaining hypotheses, H2 and H3, are tested via multivariate linear

regression in Chapter 4. Under guidance from the theoretical framework, each hypothesis

will test three models to predict variance with the variables introduced here. Models will

include a set of variables as predictors, ranging between two and nine each, and the

significance of the optimal combination will test the hypothesis. These inquiries seek to

identify niches within dementia care dyads where social disparities are most evident.

67
 CHAPTER IV

RESULTS

This chapter reviews the results of the data analysis. Descriptive analysis for primary

variables is presented, followed by discussion of how findings impact the hypotheses.

DESCRIPTIVE STATISTICS

The social aspects of dementia care are numerous and complex. Table 2 shows the

distribution of demographics and care traits among dementia caregivers and confirms much

of what NAC/AARP and NAC/Alzheimer’s Association publications (Hunt et al. 2015;

Hunt et al. 2017) have disclosed about this sample. While the distribution of race closely

resembles the make-up of the United States, caregiver age skews in favor of older caregivers

and gender demonstrates a clear skew in favor of women (although men are more

represented in dementia care than in caregiving at large). Education is split nearly down the

middle; income skews in favor of households making more than $50,000 a year, though at

this level we cannot say it is necessarily by much.

Relationships show to be an important factor in dementia care. Over half of

dementia care is between a parent (or parent-in-law) and adult child, with care of partners

representing another quarter. More caregivers support a person who is not related to them

than care for siblings, grandparents, and other relatives combined. Of dementia caregivers,

one in four has a child under 18 living with them (i.e., they constitute part of a sandwich

generation).

68
 Table 2. Independent and Intervening Variables Used in Analysis of
Social Aspects of Dementia Care Dyads, 2015 (NAC/AARP)
Variable Category (n) Sub-Category % Std. Dev. Skew
(n or mean)
Gender Women (150) 57.8 .49480 -.319
Men (109) 42.2 .49480 .319
Race White (172) 66.4 .47325 -.698
Person of Color (87) 33.6 .47325 .698
Black (35) 13.4 .49239
a
.423
a

Non-Black POC (52) 20.2 .49239

a
-.423
a

Education HS or Less (134) 51.7 .50069 -.067

No HS Degree (18) 7.0 .25609 3.382
Certified beyond HS (125) 48.3 .50069 .067
Income Above 50K (149) 57.6 .49515 -.309
Below 50K (110) 42.4 .49515 .309
Caregiver Age Over 50 (168) 64.8 .47865 -.621
18-49 (91) 35.2 .47865 .621
Relationship Family (228) 87.8 .32795 -2.323
Parent-Child (138) 53.2 .49997 -.127
Partnered (63) 24.5 .43076 1.194
Other Relative (27) 10.2 .58025 -.009
Non-Family 12.2 .32795 2.323
Sandwich 26.2 .44051 1.090
Employment Employed (161) 62.2 .48584 -.506
Not Employed (98) 37.9 .48584 .506
Scope Primary (161) 58.1 .49430 -.332
Solo (98) 37.7 .48553 .511
Unpaid (162) 62.3 .48553 -.511
Age of Care Recipient (76.6) 14.54000 -1.179
at Onset (est.) (72.4) 16.58402 -1.583
Care before 65 22.6 .41873 1.321
a
calculated within parent category only

Employment is also a significant phenomenon for dementia care. While over a third

of dementia caregivers do not work, among those who do nearly two out of three are

working at least 35 hours per week. Among primary dementia caregivers, more are operating

without help from friends or family (37.7 percent) than are (20.5 percent); this indicates that

studies excluding secondary caregivers may leave out two out of every five informal roles in

dementia care. Dementia caregivers report that one in five of their loved ones requires

constant supervision.
69
Hypothesis 1

The first hypothesis from Chapter 2 states:

H1: The age of dementia onset among care recipients varies by race.

Recall that early onset dementia has a distinct pathological trajectory from those dementias

that are recognized mostly beyond age 65. Table 3 shows the challenges discussed above

with simplistic pathological determinants that ignore social phenomena, where age of onset

appears to vary across race. H1 can be explored (if not fully confirmed) using descriptive

statistics. By comparing means across three categories of race (White, Non-Black Person of

Color, and Black), age of care recipient and, by extrapolation, age of onset, reveal themselves

to be much lower for people of color than for whites. The average white dementia care

recipient is over seven years older (79.2) than the average person of color (about 71.5), and

this disparity grows by another year when comparing age of onset (means: 75.1 for White

and 67.0 for Persons of Color). Even though each of these race categories demonstrates

about the same proportion in the dementia subpopulation as they do in the U.S. population,

the dementia populations for people of color are far younger.

Table 3. Medians for Age of Onset, Controlled for Race, 2015 (NAC/AARP)
Variable White Non-Black POC Black
(N) (172) (52) (35)
mean Std. Dev. mean Std. Dev. mean Std. Dev.
Age of Care Recipient 79.2 13.43800 71.6 16.05000 71.5 14.55700
Age at Onset (est.) 75.1 14.95530 67.1 19.37383 67.0 16.96655
Early Onset 16.1% .36855 36.5% .48605 34.5% .47908

When these figures are graphed, both the starkness of the distribution and its

methodological limitations are apparent. Figure 1 shows the distribution of dementia

70
cases by calculated age of onset, controlling for race. The distribution curve for whites is the

most robust, as it has the most cases, but it also distinctly crests far to the right of (higher

than) Age 65, the benchmark for early onset dementia. The distribution curves for Black and

non-Black people of color, on the other hand, crest very close to Age 65. These curves are

rather flat due to their smaller number of cases. The significance of these findings must be

tested via bivariate correlation.

Figure 1. Age of Onset, Dementia/Memory Problems (NAC/AARP 2015)

BIVARIATE CORRELATIONS

Measuring the significance of a relationship between categorical variables and linear

variables calls for cross-tabulations; cross-tabulation measures the significance of

relationships by measuring correlation between all categories of variables analyzed. Age of

onset has too few cases spread across too many categories to compare as an interval-ratio

variable, but its reduction to the early onset dichotomous variable can be utilized instead.

Table 4 reports the results of this comparison, confirming a significant, weak relationship (χ²

71
= 12.302, Φ = -0.218 and Cramer’s V = 0.218). H1 is confirmed through controlled

descriptive statistics. The age of dementia onset among care recipients is higher among

people of color than among whites. Since nearly half of all cases of dementia for people of

color are diagnosed before age 65, the early onset benchmark seems to conflate experiences

that are not, in fact, the same.

Table 4. Cross-Tabulation of Age of Onset

(Dementia/Memory Problems) by Race, 2015 (NAC/AARP)
Variable All Dyads White Persons of Color
Early Onset
No 77.4% 83.7 64.4%
Yes 22.6% 16.3 35.6%
Total 100% 100% 100%
N 259 172 87
χ²=12.302, df=1, p≤0.001
Φ=-0.218 and Cramer’s V=0.218, p≤0.001

Before proceeding to regressions, the strength of other relationships related to this

study’s remaining hypotheses can be tested. Many have not demonstrated a significant

relationship with the outcome variables, but those that do are reported in Table 5. Some

moderate linear relationships are evident, notably the negative relationships employment

status demonstrates with outcomes when stratified by a demographic; that is, non-Black

persons of color who are employed tend to report significant lower levels of physical stress;

also, an observed reduction in self-reported emotional stress for caregivers who are

employed appears to be concentrated in women.

Controls reveal concentrations of relationship within specific categories. A negative

relationship between employed caregivers and physical strain, for example, appears to be

concentrated in caregivers who are neither white nor Black. As discussed in the literature

review, research on non-Black people of color (recorded as Black, Hispanic, Asian, or Other)

72
is limited but has repeatedly suggested that communities of color have a strong sense of

family duty that may mitigate stressors noted in general care populations. This assertion is

especially common for Black caregivers, and yet they lack a significant relationship (even a

negative one) here. Similarly, a negative relationship between employment and emotional

stress appears to be concentrated in women. In both cases, the correlation coefficient may

reflect social role fulfillment (communities of color fulfilling cultural expectations of family

care, women who are relieved to have responsibilities outside the home) or it may reflect a

low requirement of care: a caregiver who is able to work may not be overwhelmed by

responsibilities at home, or they may have plenty of unpaid support to share the burden.

Table 5. Correlation Matrix of Select Caregiver Demographics and

Outcomes, 2015 (NAC/AARP)
Self-Reported Self-Reported ADL
Physical Strain Emotional Count
(5-Level) Stress (5-Level) (0-7)
Std. Dev. 1.26160 1.17694 2.13535
Demographics
Woman .49480 -.099
+Employed -.225**
White .47325 .055 -.133*
+Recipient Under 65 -.174*
Non-Black Person of Color
+Employed -.329*
Care Traits
Recipient Under 65 .38124 -.180**
Parent-Child .49997 .052 .181**
Sandwich .44051
Employed .48584 -.154*
* Correlation is significant at the .05 level (2-tailed).
** Correlation is significant at the .01 level (2-tailed).

Note that these findings do not rule out a relationship between variables, only

confirm the strength and direction of straightforward linear relationships. Table 5 shows

only significant relationships that are unstratified or controlled for a single intervening

73
variable. Exploring more than two variables at a time (or three variables, if one is a control)

requires a multivariate methodology, such as regression.

MULTIVARIATE LINEAR REGRESSIONS

Multivariate linear regression measures the strength and direction of relationships

between interval-ratio variables, dichotomous variables, or some combination thereof, using

ordinary least squares to approximate a normal distribution and predict variance for a

dependent variable. The remaining two hypotheses will be tested for significance through

three models each of regression; if multiple models are found to be significant, the most

elegant model (highest variance predicted with fewest variables) will be the most effective.

Hypothesis 2

The second hypothesis from Chapter 2 states:

H2: Dementia caregivers with normative characteristics will demonstrate less physical strain
than non-normative caregivers.

For dementia care dyads, the following traits are so common that their overlap may

constitute a normative niche within dementia care: white, woman, over 50, higher income,

not employed, caring for a parent. If the niche is significant, it could skew results for all non-

normative caregivers. This hypothesis will be confirmed if some combination of the

variables listed demonstrates a negative relationship with physical strain, but the higher the

R² score, the more impactful the niche.

Results from three models are reported in Table 6. Model 2a is the simplest: what is

the strength of the relationship between white woman and physical strain in dementia care?

Model 2b includes all six of the variables listed to determine whether the niche gains

significant accuracy with all factors included. Tweaking from among these six variables did
74
not improve significance much, so three further dichotomous variables are added for Model

2c: less education (high school diploma or less), normal onset (i.e., dementia care begins at or

after age 65), and unpaid support. These three variables showed some significance when

added to Model 2b individually, but their cumulative impact was strongest.

Table 6. Regression Models Predicting Normative Respondents’

Self-Reported Physical Strain, NAC/AARP 2015
Model 2a Model 2b Model 2c
b β b β b β
Constant 2.010 2.025 2.117
(Standard Error) (.163) (.236) (.291)
White -.264 -.099 -.308 -.116 -.262 -.098
(.165) (.168) (.172)
Woman -.168 -.066 -.176 -.069 -.172 -.068
(.158) (.158) (.158)
Over 50 -.193 -.073 -.206 -.078
(.173) (.173)
HS or Less -.124 -.049
(.169)
Higher Income -.203 -.080 -.253 -.099
(.161) (.177)
Not Employed .280 .108 .310 .119
(.166) (.168)
Caring for a Parent .182 .072 .191 .076
(.160) (.165)
Normal Onset -.209 -.069
(.200)
Unpaid Support .189 .073
(.166)
R² .014 .044 .054
F 1.852 1.920 1.573
N 259 259 259
No result met the (1-tailed test) significance threshold of p≤0.05.

This demonstrates that complex social factors (including, intersectionality would note,

oppression) not only build upon one another in small, individual increments, they can also

compound for an even greater effect than the sum of their parts. By adding more normative

variables, we may get closer to demonstrating the hypothesis. Unfortunately, this sort of

nickel-and-dime approach is discouraged in statistics — as the degrees of freedom increase,

75
the broad applicability of findings diminishes — and for this reason none of the models met

the significance threshold.

Having failed to reject the null hypothesis, no normative niche is confirmed for this

population of dementia caregivers. No significant relationship between normative caregivers

and physical strain was observed. It appears that an analysis centered on normative social

forces is not effective for demonstrating intersectional and othering dynamics among

dementia caregivers.

Hypothesis 3

The third hypothesis from Chapter 2 states:

H3: Dementia caregivers with higher social marginalization will demonstrate a higher ADL
index than other caregivers.

To demonstrate that social forces from outside caregiving can still affect caregivers, some of

the starkest social factors need to be discussed in their caregiving context. No studies have

explored racism, sexism, or income inequality in the context of care dyads (although some

reviews discuss homophobia in the context of LGBTQ caregivers, and as discussed above

health research in general is embracing complex social and life course considerations), but

these forces are inescapable because caregiving does not happen in a vacuum. Forms of

discrimination and systemic oppression interact with one another, and intersectional analysis

allows us to explore how. This test looks at ADL index, one of the most quantitative

measures available for caregiver burden. To test whether the measure of multiple

oppressions can be significantly predictive in the context of dementia care, disadvantaged

demographic categories are tested: person of color, woman, lower income, less education,

76
employed, and sandwich generation. This hypothesis must demonstrate statistical

significance before its measure of strength becomes relevant.

Table 7 reports the results of the three models tested for H3.

Table 7. Regression Models Predicting Marginalized Respondents’ Index of

Activities of Daily Living Supported, NAC/AARP 2015
Model 3a Model 3b Model 3c
b β B β b β
Constant 2.010 2.025 2.117
(Standard Error) (.163) (.236) (.291)
Person of Color .601 .133 .631 .139 .599 .132
(.279) (.284) (.283)
Woman -.183 -.042 -.206 -.048 -.219 -.051
(-.267) (.270) (.269)
Below 50K .441 .102 .334 .077
(.293) (.282)
HS or Less .040 .009
(.289)
No HS Diploma .874 .102
(.552)
Employed -.277 -.063 -.271 -.061
(.279) (.276)
Sandwich -.356 -.073 -.357 -.073
(.308) (.307)
R² .020 .042 .051*
F 2.557 1.811 2.244
N 259 256 256
*p≤0.05 (1-tailed test)

Model 3a tests the significance of women of color’s relationship with ADL index. Model 3b

tests all seven variables listed at once. Neither demonstrates a significant relationship, yet

Model 3b is close (its significance measured just outside the standard p-value for regressions,

p ≤ 0.05). In this model, education was a surprisingly weak variable (far lower than in Model

2C); as stated above, the dichotomous split around high school diploma is very close to fifty-

fifty, potentially offering too little variance to be significant. A new variable was recoded

from the original dataset: a dichotomous measure for caregivers who had the least education

(no high school diploma). Model 3c replaced the lower education variable for least education

77
and identified a significant relationship. In Model 3c, education becomes the most impactful

predictor, alone accounting for variance of nearly one (0.874) ADL point — not a small

difference for a variable with only a range of 6 and a mean of 2.1. This indicates that, among

the most marginalized dementia caregivers (women of color with lower income who are

employed and part of a sandwich generation), not having a high school diploma raises by

almost one the number of ADLs performed daily.

It is important to clarify what the finding implies. It does not necessarily follow that

the conjunction of some-but-not-all caregivers with marginalized predictors will experience

the same impact or even the same kind of impact — intersectional interactions are

contextual. A person without a high school diploma is more likely than a person with a

diploma to come from a family facing tremendous disadvantage, socially and health-wise, so

their loved ones tend to need much more from them on average. There is not likely to be

any direct or indirect relationship between a caregiver’s high school diploma and a present

care recipient’s care needs, but it might offer insight into how caregivers are selected.

It does not appear children from smaller families (therefore fewer available

caregivers) attain less education than those with many siblings (in fact, the opposite seems

more likely; Falbo & Polit 1986); more likely, a family who is choosing the caregiver for an

elder with dementia may favor, among those available, persons with if not the least

education, then the meagerest income or job prospects or the fewest meaningful

accomplishments in the eyes of their family. Whatever an individuals’ personal reason for

not finishing high school (“deliberately, accidentally, or even of necessity,” to quote

Halberstam), by not pursuing the national baseline for education, the person may signal that

78
making more money is not a priority; such a person may live “outside the logics of capital

accumulation” and thus be regarded by their family as one of Halberstam’s queer subjects

(2005:10). In terms of economic contribution, a person who has very little education may be

regarded non-normative, increasing the likelihood their family would thrust into another

queer (subjectivity) role: that of caregiver.

Other disadvantages compound for even stronger effects, indicating that caregiving

is susceptible to social and systemic marginalization. The possibilities warrant further

consideration.

79
 CHAPTER V

CONCLUSION

SUMMARY

This quantitative study presents a path forward for informal dementia care research

in the United States. As demonstrated in the literature review, caregiving and dementia care

have seen social inquiry languish while diffuse studies revisit pathology, demographics, and

micro-interactions without coordination or innovation; this study makes a case for stronger

centralization and utilization of theory and disciplinary frameworks in general and

sociological and critical approaches specifically.

To demonstrate some of the practices here endorsed, this study performed an

exploratory, quantitative analysis of social marginalization within a secondary dataset. Out of

three hypotheses, two were confirmed: that age of dementia onset among care recipients

varies by race (H1) and that dementia caregivers who are highly marginalized socially report

performing more ADLs for their care recipients (H3). The other hypothesis, that a normative

caregiver will report lower levels of physical strain, failed to attain significance. These

hypotheses were informed by Halberstam’s queer subjectivity, which asks if anyone who is

not actively procreating and building capital can be othered in society, even if they are for

reasons of family. These hypotheses were further developed through an intersectional lens to

show that dementia caregivers do not exist outside the world of the social (and particularly

of discrimination and systemic disparity), but instead have their care experiences affected by

social factors outside the care dyad. These findings should inform future research and policy

80
priorities as the U.S. and other post-industrial economies approach a surge in elder

populations and face new healthcare challenges in the decades ahead.

DISCUSSION

Western philosophy can be historically located as a discourse that accompanies

capitalism, and works to justify and rationalize a patently brutal and unjust system as
inevitably scientific and organic. So seamlessly has capitalism been rationalized over
the last two hundred years, in fact, that we no longer see the fault lines that divide
black from white, work from play, subject from object.
Judith (Jack) Halberstam
(summarizing Lindon Barrett’s Blackness and Value: Seeing Double)
In a Queer Time & Place, 2005, 8-9

One does not have to subscribe to the post-positivistic critical theories of

Halberstam or Barret or Foucault to perceive that the caregiver is both everywhere and

nowhere in society. As thousands of generations have done before, families are stepping up

every day to tend to loved ones who can no longer care for themselves, and yet society does

not normalize caregiving or caregivers. Perhaps worse is dementia care; caring for someone

who cannot recover and may one day forget you does not build families or make the family

any money. Caregiving is an action outside of normative time and space — what is a bedtime

when your loved one is sundowning? where is home when someone who once comforted

you gets lost in their own body? are we allowed to care about things and people and

relationships that society does not make an explicit time and space for, and if so how? This

study does not ask those questions overtly, but they are present. If caregiving is going to

meet the needs of elders in the coming decades, some of these tensions between the seen

and the unseen, the valued and the unvalued, will require reconciliation.

Discrimination can include erasure as well as violence, but unlike homophobia,

racism, and sexism, there is no term for the exclusion of people who prioritize care over
81
capital. Like many systemic otherings, it does not depend on the violent acts of individuals

who wish harm, only broad inaction or negligence, facilitated by a general lack of inclusion,

no normative time and space where one may be a caregiver and still be regarded as

normative in society. This is especially true for dementia caregivers, whose support happens

on a scale of years, not weeks, and who spend at least some portion of their time

accommodating a fluid, subjective reality that may exist for no one else. Through the lens of

Halberstam, caregiving emerges as queer subject, a (paradoxically social, in that it is the act

of supporting another person) social deviant.

Informal care exists outside of the economy because there is no place for it within

the economy, and yet that economy depends on free care labor. Its place is unwelcome,

othered, queer, deviant. Caregiving is not inherently anti-capitalist, and yet the evidence

presented here requires we ask whether capitalism might be anti-caregiving. If this is the

case, caregivers would find themselves in good company; at the heart of capitalism as we

know it is a hegemony that systemically (and sometimes overtly) disfavors women, people of

color, workers, LGBTQ folks, asexuals, non-monogamists, couples who are childless by

choice, and anyone who does not measure quality of life in dollars, years, or number of lives

given, taken, or saved. At the heart of these others is a sense of mutualism, either because

that is what capitalism rejects about them or in defense against the rejection. Is the caregiver

— isolated, choiceless, sacrificing, yet steadfast — all that different?

The critical literature review presented supports developing standards and

collaboration in care research and demonstrates the practice through a quantitative inquiry.

Scholarship up to this point has been so diffuse one could call it anti-disciplinary, but by a

82
lack of intention rather than a deliberate deconstruction. Because this study has also

acknowledged the relevance of post-positivistic approaches, it raises epistemological and

ontological questions about creating new academic structures while accommodating post-

structuralist knowledge and even priorities. In approaching this study with critical theories,

my intention was to not overlook the process of othering in our social worlds; too many

fields of research were built first and made accessible much later, but caregiving could be

different. Caregiving, as much as sociology itself, is about people and for people, but can a

foundation be built that leaves room for future discoveries and corrections?

Queer subjectivity empowers us to consider the possibility that caregiving is a

transgression, and intersectionality tells us that transgressions face social punishment with

material consequences. The theories employed in this study were selected because any

mistakes of their application will be more informative than the correct application of

uncritical social science. If it is found that the author has misappropriated queer subjectivity

or Black feminism, then let the corrections be fluent and accessible and waste no time in

creating space for caregivers, who may not even see themselves well enough to know they

are no longer normative.

The quantitative analysis presented here is only a step in the direction of uncovering

the full, non-normative mechanics of informal care. Many dementia caregivers face a

powerful separation between their inner and outer inner worlds, but understanding of their

lived experience can be improved if we expand the scope of inquiry to include the networks

of support, the selection of individual caregivers by families, and the dense interdependence

of the act of care itself. Demonstrating the commonalities within and between types of

83
caregivers (not only by condition, but also by social and cultural contexts) can reinvigorate

scholarship but also deepen the connectivity of caregivers between themselves, giving them

voice and their experiences validity.

H1 confirms that caregiving, pathology, and social factors affect one another and

warrant consideration together. The mean disparity in ages of onset between whites and

people of color is severe enough, we are left to ask whether 65 is all that early for people of

color; since early onset dementia has its own pathological trajectory, how can we know

whether people of color are receiving the most appropriate medical response? The

determination of early onset dementia may constitute a mechanism of systemic racism

affecting healthcare and caregiving and warrants further examination. Critical theories beg us

to go one step further and ask how the universal demarcation of age 65 contributes to the

othering of people of color with dementia and therefore how it serves white supremacy.

The rejection of H2 discounts the notion of a normative dementia caregiver based on

the most common caregiver demographics. This finding remains instructive about non-

intersectional layering of identities: it is not enough to be most common; identity layering in

research must be linked to specific (usually oppressive) social forces.

The findings for H3 return this inquiry to its intersectional perspective and confirm

that, though small in this model, dementia caregivers can show marginalization in their work

and/or selection. H3 expands evidence from H1 that social disparities interact with care

circumstances. Unlike the pathological lens for H1 (i.e., the age that determines whether

dementia is early onset or not), Model 3c presents only social relationships as predictors:

84
how is the person racialized and genderized, what is their economic and educational access,

what responsibilities do they hold besides care.

This study confirms two hypotheses and fails to confirm one. H1 confirms that the

pathology of dementia is affected in some way by racialization. H3 confirms that social

determinants can directly correlate to care needs on the ground. The social lives of caregivers

are significant to the course and outcomes of dementia care.

The prominence of caregivers seems poised to rise further as elder and disabled

populations come to represent a more significant segment of society; however, such an

emersion challenges normative family and economic priorities. Scholars cannot deny that

care dynamics demonstrate social influence, nor that quantitative research benefits from

consideration of intersectional compounding of normative/non-normative othering. Nor

can dementia caregivers see their circumstances only through the lens of one sick relative:

the circumstances of dementia care are just as systemic as they are personal.

LIMITATIONS

By emphasizing that new truths available in a secondary source, this study forgoes

the opportunity to ask further questions from participants or researchers involved. Its

emphasis on dementia care will also limit the reliability of results for other types of

caregivers. Findings will only be directly applicable to experiences within the United States

and may not accurately reflect trends in other countries.

The design of this study is nascent and requires further development. That said,

these choices have been made deliberately to specifically highlight gaps in existing literature,

85
outline opportunities for methodological innovation, and more effectively elucidate the

social lives of informal dementia caregivers.

RECOMMENDATIONS

Developing a stronger sociological base for care research will require the

development of both knowledge and design; before caregiving can begin to consider

standardization under sociology or some other field, some cohort of investigators will have

to agree that this is worthwhile and commit to building bridges between the research niches

that currently exist. These bridges will require social scientists to become more familiar with

biomedical scholarship (and vice versa) and to improve social inquiry of the middle ground

between the four farthest corners of care scholarship: hyperlocal qualitative, logistical

biomedical practice, aloof quantitative demography, and abstract theory. Researchers may

also develop interdisciplinary or trans-disciplinary research that links caregiving trends with

links to relevant inquiries in other fields, such as generosity (anthropology), non-romantic

interdependence (psychology), and mutual aid (political science). These approaches all have

something to show one another, but without networking across all levels care research will

remain anti-structural and anti-disciplinary.

While building a framework for care research that crosses levels and disciplines could

help to fill research gaps, some of the limitations could be improved by implementing more

theoretical approaches that contextualize individual care dynamics in greater economic and

power trends. By exploring the concept of normativity, I attempted to place the positivistic

in conversation with the post-positivistic, but such an undertaking will require a larger

dataset or tighter construct of normativity to gain significance. Researchers would do well in

86
such an undertaking to make it clear that defining normativities creates others out of the

non-normative; though the outcomes are vastly different, there is a fine line between

detailing existing disparities and reinforcing them.

The health of greater families and support systems cannot be discerned from any

available caregiver dataset, but the present findings bolster the case for more comprehensive

inquiry of whole networks and systems around dementia care, not only the primary dyad.

Understanding the dyad requires a more systemic reflection on the dyad’s familial support

and circumstances. However, even in available datasets, these findings can be expanded by

controlling for family versus non-family care dyads (particularly H3).

Both theoretical frameworks employed here have stronger origins in interpretation

than in quantitative analysis. This confluence of methods may be off-putting to readers who

prefer one ontology or another — positivism, or post-positivism — but it should

nonetheless inform future endeavors in social theory, quantitative intersectionality, and

mixed methodologies.

The techniques utilized here found weak significant results; future studies may

consider linking intersectional approaches with other multivariate techniques, such as factor

analysis, cluster analysis, and try to improve upon these findings. Although the NAC/AARP

dataset was selected for its breadth of both demographic and subjective variables, this study

warrants repeating with a larger (probably federal) dataset, such as the National Survey of

Caregivers. Politicization affects what research is performed and shared from the federal level;

non-federal secondary datasets can aid in filling potential gaps, but would be bolstered if

87
they were more normalized in academic literature and their range and complexity taught,

assessed, and evaluated as part of graduate level social science education.

Research into how dementia is framed in popular culture may help to alleviate the

disconnect and elevate the agency of people in early stages of dementia and the humanity of

those in late stages. For all the reminders in dementia care scholarship that dementia is not a

mental illness, it is still socialized like one. Such research may want to invoke the literature of

disability studies, particularly calls to amplify the voices of disabled people in defining their

own experiences.

Finally, and perhaps most importantly, researchers need to incorporate a stronger

sense of stakeholder in studies of caregiving. The researcher and the medical practitioner

offer important insights but can still overlook elements of care that may be more apparent to

caregivers, their extended support system, and even persons with dementia themselves. Only

by including multiple (and, ideally all) of these perspectives will care research advance

beyond its heretofore reactive stance.

CONCLUSION

As this study was being completed, the world took on great, collective, health-

anxious gasp as a new and highly contagious virus transversed every populated continent.

While epidemiologists go to work to inform the public and mitigate the impact, topics of

mutual care and social isolation are suddenly prominent and urgent. Where those

conversations lead remains to be seen, but somewhere between the panic and the boredom

there have also been indications that not everyone who beats COVID-19 is able to fully

recover. It is too early to tell what the impact will be upon the elder population — a high

88
enough death rate for people over 65 could completely upend predictions of a “care gap” —

but two other populations are changing in demographically significant ways. Firstly, in the

United States, elders of color (especially Black men) have been the most likely to die in the

first wave of COVID-19, as decades of continued health disparity and systemic disadvantage

take their toll. Secondly, there is a growing likelihood that some portion of COVID-19

survivors will become permanently disabled in ways that have not yet been identified,

creating a new care recipient population of unknown scope.

Caregiving has always been relevant, but never more urgent than right now. How we

care for others and expect to be cared for may set the stage for the next century of social

habits, expectations, and connections. From dementia caregivers, we can learn to accept a

reality that is not entirely our own, to recognize and respect the agency of others, and to

develop and sustain strong support systems — informal and formal — that serve all

participants well.

89
 REFERENCES

Adler, Richard, and Rajiv Mehta. 2014. Catalyzing Technology to Support Family Caregiving.
Washington, DC: National Alliance for Caregiving.
Angel, Jacqueline L., Stipica Mudrazija, and R. Benson. 2015. Racial and Ethnic Inequalities in
Health. Oxford: Elsevier Science & Technology.
Apesoa-Varano, Ester C., Judith C. Barker, and Ladson Hinton. 2011. "Curing and Caring:
The Work of Primary Care Physicians with Dementia Patients." Qualitative Health
Research 21(11):1469-1483.
Apesoa-Varano, Ester C., Yajarayma Tang-Feldman, Susan C. Reinhard, Rita Choula, and
Heather M. Young. 2015. "Multi-Cultural Caregiving and Caregiver Interventions: A
Look Back and a Call for Future Action." Generations: Journal of the American Society on
Aging 39(4):39-48.
Bailes, Christine O., Colleen M. Kelley, and Nadine M. Parker. 2016. "Caregiver Burden and
Perceived Health Competence when Caring for Family Members Diagnosed with
Alzheimer's Disease and Related Dementia." Journal of the American Association of Nurse
Practitioners 28(10):534-540.
Barnes, Marian, and Tula Brannelly. 2008. "Achieving Care and Social Justice for People
with Dementia." Nursing Ethics 15(3):384-395.
Bauer, Greta R., and Ayden I. Scheim. 2019. "Advancing Quantitative Intersectionality
Research Methods: Intracategorical and Intercategorical Approaches to Shared and
Differential Constructs." Social Science & Medicine 226:260-262.
Beard, Renée L., and Patrick J. Fox. 2008. "Resisting Social Disenfranchisement: Negotiating
Collective Identities and Everyday Life with Memory Loss." Social Science & Medicine
66:1509-1520.
Beard, Renée, Jenny Knauss, and Don Moyer. 2009. "Managing Disability and Enjoying
Life: How We Reframe Dementia through Personal Narratives." Journal of Aging Studies
23:227-235.
Bédard, Michel, D. W. Molloy, Larry Squire, Sacha Dubois, Judith A. Lever, and Martin
O'Donnell. 2001. "The Zarit Burden Interview: A New Short Version and Screening
Version." The Gerontologist 41(5):652-657.
Berry, Brandon, and Ester C. Apesoa-Varano. 2017 “Medication Takeovers: Covert
Druggings and Behavioral Control in Alzheimer’s.” Social Science & Medicine 188:51-59.
Berry, Brandon, Ester C. Apesoa-Varano, and Yarin Gomez. 2015. "How Family Members
Manage Risk Around Functional Decline: The Autonomy Management Process in
Households Facing Dementia." Social Science & Medicine 130:107-114.

90
Bookman, Ann, and Delia Kimbrel. 2011. "Families and Elder Care in the Twenty-First
Century." The Future of Children 21(2):117-140.
Bosco, Alessandro, Justine Schnieder, Donna Maria Coleston-Shields, and Martin Orrell.
2019. “Dementia Care Model: Promoting Personhood through Co-Production.”
Archives of Gerontology and Geriatrics 81:59-73.
Brodaty, Henry, and Marika Donkin. 2009. "Family Caregivers of People with Dementia."
Dialogues in Clinical Neuroscience 11(2):217-228.
Bruhn, John G., and Howard M. Rebach. 2014. The Sociology of Caregiving. New York:
Springer.
Caceres, Billy A., Mayu O. Frank, Jin Jun, Melissa T. Martelly, Tina Sadarangani, and Paloma
C. de Sales. 2016. "Family Caregivers of Patients with Frontotemporal Dementia: An
Integrative Review." International Journal of Nursing Studies 55:71-84.
Cloyes, Kristin G., William Hull, and Andra Davis. 2018. "Palliative and End-of-Life Care
for Lesbian, Gay, Bisexual, and Transgender (LGBT) Cancer Patients and their
Caregivers." Seminars in Oncology Nursing 34(1):60-71.
Cockerham, William C. 2013. "The Social Causation of Health and Disease." Pp. 1-26 in
Social Causes of Health and Disease. Oxford: Polity Press.
Cockerham, William C. and Ferris J. Richey. 1997. “Introduction: An Overview of Medical
Sociology.” Pp. xi-xxvi in Dictionary of Medical Sociology. Westport, Connecticut:
Greenwood Press.
Cook, Sarah K., Lauren Snellings, and Steven A. Cohen. 2018. "Socioeconomic and
Demographic Factors Modify Observed Relationship between Caregiving Intensity and
Three Dimensions of Quality of Life in Informal Adult Children Caregivers." Health and
Quality of Life Outcomes 16:169.
Cornwell, Benjamin, and Markus Schafer. 2015. "Social Networks in Later Life." Pp. 181-
201 in Handbook of Aging and the Social Sciences. Oxford: Elsevier Science and Technology.
Crenshaw, Kimberlé. 1991. "Mapping the Margins: Intersectionality, Identity Politics, and
Violence Against Women of Color." Stanford Law Review 43(6):1241-1299.
------. 1993. "Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of
Antidiscrimination, Feminist Theory and Antiracist Politics." University of Chicago Legal
Forum:139-167.
Crimmins, Eileen M., and Sarinnapha M. Vasunilashorn. 2016. "Biodemography: Adding
Biological Insight into Social, Economic, and Psychological Models of Population and
Individual Health Change with Age." Pp. 55-75 in Handbook of Aging and the Social
Sciences., edited by L. George, and K. Ferraro. Oxford: Elsevier Science & Technology.
Dam, Alieske E. H., Marjolein E. de Vugt, Inge P. M. Klinkenberg, Frans R. J. Verhey, and
van Boxtel, Martin P. J. 2016. "A Systematic Review of Social Support Interventions for

91
 Caregivers of People with Dementia: Are They Doing What They Promise?" Maturitas
85:117-130.
Davey, Adam, and Maximiliane E. Szinovacz. 2008. "Division of Care Among Adult
Children." Pp. 133-159 in Caregiving Contexts: Cultural, Familial, and Societal Implications.
New York: Springer.
DeNavas-Walt, Carmen, and Bernadette D. Proctor. 2015. Income and Poverty in the United
States: 2014. Washington, DC: U.S. Census Bureau. Accessed May 11, 2020.
https://www.census.gov/content/dam/Census/library/publications/2015/demo/p60-
252.pdf
Dunham, Charlotte, and Julie Cannon. 2008. "‘They're Still in Control Enough to Be in
Control’: Paradox of Power in Dementia Caregiving." Journal of Aging Studies 22(1):45-53.
Dwolatzky, T., and A. M. Clarfield. 2003. "Correspondence: Reversible Dementias." Journal
of Neurology, Neurosurgery, and Psychiatry 74:1008-1010.
Edelman, T. 2018. "Buying and Selling Nursing Homes: Who's Looking Out for the
Residents?" Accessed September 6, 2019. http://medicareadvocacy.org/buying-and-
selling-nursing-homes-whos-looking-out-for-the-residents
Eifert, Elise K., Rebecca Adams, Sharon Morrison, and Robert Strack. 2016. "Emerging
Trends in Family Caregiving using the Life Course Perspective: Preparing Health
Educators for an Aging Society." American Journal of Health Education 47(3):176-197.
Evans, Clare R. 2019. "Adding Interactions to Models of Intersectional Health Inequalities:
Comparing Multilevel and Conventional Methods." Social Science & Medicine 221:95-105.
Falbo, Toni, and Denise F. Polit. 1986. "Quantitative Review of the Only Child Literature:
Research Evidence and Theory Development." Psychological Bulletin 100(2):176.
Fazio, Sam, Douglas Pace, Janice Flinner, and Beth Kallmyer. 2018. "The Fundamentals of
Person-Centered Care for Individuals with Dementia." The Gerontologist 58(S1):S10-S19.
Feinberg, Lynn F., and Brenda C. Spillman. 2019. "Shifts in Family Caregiving — and a
Growing Care Gap." Generations: Journal of the American Society on Aging 43(1):73-77.
Ferrer, Ilyan, Amanda Grenier, Shari Brotman, and Sharon Koehn. 2017. "Understanding
the Experiences of Racialized Older People through an Intersectional Life Course
Perspective." Journal of Aging Studies 41:10-17.
Flinn, Brendan. 2018. Millennials: The Emerging Generation of Family Caregivers. Washington, DC:
AARP Public Policy Institute.
Foucault, Michel. 1994. The Birth of the Clinic: An Archaeology of Medical Perception. Translated by
A.M. Sheridan Smith. New York: Vintage.
Fox, Susannah, Maeve Duggan, and Kristen Purcell. 2013. Family Caregivers Are Wired for
Health. Washington, DC: Pew Research Center. Accessed December 24 2018.
http://www.pewinternet.org/Reports/2013/Family-Caregivers.aspx

92
Frank, Adam. 2011a. "Beyond the Punch-Clock Life: The Tyranny of Modern Time II."
Accessed April 25, 2020. http://www.npr.org/blogs/13.7/2011/09/27/140818962/
beyond-the-punch-clock-life-the-tyranny-of-modern-time-ii.
------. 2011b. "The Future of Time." Accessed April 25, 2020.
http://www.npr.org/sections/13.7/2011/10/12/141217278/the-future-of-time.
------. 2011c. "The Tyranny of Modern Time." Accessed April 25, 2020.
http://www.npr.org/blogs/13.7/2011/09/23/140718434/time-crisis-why-you-don-t-
care-about-today-s-equinox.
Fraser, Nancy. 2016. "Contradictions of Capital and Care." New Left Review 100(Jul/Aug):99-
117.
Friedemann, Marie-Luise, and Kathleen C. Buckwalter. 2014. "Family Caregiver Role and
Burden Related to Gender and Family Relationships." Journal of Family Nursing
20(3):313-336.
Fry, Richard. 2017. "5 Facts about Millennial Households." Accessed August 8, 2019.
https://www.pewresearch.org/fact-tank/2017/09/06/5-facts-about-millennial-
households/
Gaugler, Joseph, Bryan James, Tricia Johnson, Allison Marin, and Jennifer Weuve. 2019.
2019 Alzheimer's Disease Facts and Figures. Chicago: Alzheimer’s Association.
Gauthier, S., A. Leuzy, E. Racine, and P. Rosa-Neto. 2013. "Diagnosis and Management of
Alzheimer's Disease: Past, Present and Future Ethical Issues." Progress in Neurobiology
110:102-113.
George, Linda K., and Kenneth F. Ferraro. 2016. "Aging and the Social Sciences: Process
and Prospects." in Handbook of Aging and the Social Sciences, edited by L.K. George and
K.F. Ferraro. Oxford: Elsevier Science & Technology.
Gitlin, Laura N., Helen C. Kales, Katherine Marx, Barbara Stanislawski, and Constantine
Lyketsos. 2017. "A Randomized Trial of a Web-Based Platform to Help Families
Manage Dementia-Related Behavioral Symptoms: The WeCareAdvisor." Contemporary
Clinical Trials 62:27-36.
Gkiouleka, Anna, Timothy Huijts, Jason Beckfield, and Clare L. Bambra. 2018.
"Understanding the Micro and Macro Politics of Health: Inequalities, Intersectionality
& Institutions — A Research Agenda." Social Science & Medicine 200:92-98.
Gleckman, Howard. 2019. "Another Shock to the Long-Term Care Insurance Industry."
Accessed September 6, 2019. https://www.forbes.com/sites/howardgleckman/2019/
04/01/another-shock-to-the-long-term-care-insurance-industry/
Glenn, Evelyn N. 2010. Forced to Care. Cambridge: Harvard University Press.
Gouin, Jean‐Philippe, Chelsea Estrela, Kim Desmarais, and Erin T. Barker. 2016. "The
Impact of Formal and Informal Support on Health in the Context of Caregiving
Stress." Family Relations 65(1):191-206.
93
Green, Mark A., Clare R. Evans, and S. V. Subramanian. 2017. "Can Intersectionality Theory
Enrich Population Health Research?" Social Science & Medicine 178:214-216.
Greenwood, Nan, Gillian Mezey, and Raymond Smith. 2018. "Social Exclusion in Adult
Informal Carers: A Systematic Narrative Review of the Experiences of Informal Carers
of People with Dementia and Mental Illness." Maturitas 112:39-45.
Grusky, David B., and Katherine R. Weisshaar, eds. 2014. Social Stratification: Class, Race, and
Gender in Sociological Perspective. 4th ed. Boulder: Westview Press.
Gurrentz, Benjamin. 2018. Millennial Marriage: How Much Does Economic Security Matter to
Marriage Rates for Young Adults? Washington, DC: U.S. Census Bureau.
Guy-Sheftall, Beverly, ed. 1995. Words of Fire: An Anthology of African-American Feminist
Thought. New York: New Press.
Haeusermann, Tobias. 2017. "The Dementias — A Review and a Call for a Disaggregated
Approach." Journal of Aging Studies 42:22-31.
Halberstam, Judith. 2005. In a Queer Time and Place: Transgender Bodies, Subcultural Lives. New
York: New York University Press.
Hall, Donald E. 2003. "Introduction: What ‘Queer Theories’ Can Do for You." Pp. 9-18 in
Queer Theories. New York: Palgrave Macmillan.
Hancock, Ange-Marie. 2016. Intersectionality: An Intellectual History. New York, NY: Oxford
University Press.
Hankivsky, Olena. 2012. "Women's Health, Men's Health, and Gender and Health:
Implications of Intersectionality." Social Science & Medicine 74(11):1712-1720.
Henning-Smith, Carrie, Megan Lahr, and Michelle Casey. 2019. "A National Examination of
Caregiver use of and Preferences for Support Services: Does Rurality Matter?" Journal of
Aging and Health 31(9):1652-1670.
Hermanns, Melinda, and Beth Mastel-Smith. 2012. "Caregiving: A Qualitative Concept
Analysis." The Qualitative Report 17(Art. 75):1-18. Accessed June 25, 2019.
http://www.nova.edu/ssss/QR/QR17/hermanns.pdf
Hirschman, Karen B., and Nancy A. Hodgson. 2018. "Evidence-Based Interventions for
Transitions in Care for Individuals Living with Dementia." The Gerontologist 58(S1):S129-
S140.
Hodson, G.J. forthcoming. "Modeling Types of Informal Caregivers: Gender, Condition,
and Social Age." Gender and Generations: Advances in Gender Research, Vol. 30, edited by V.
Demos and M.T. Segal. Somerville: Emerald.
hooks, bell. 1991. "Theory as Liberatory Practice." Yale Journal of Law and Feminism
4(1):Article 2.

94
Hunt, Gail Gibson, Rick Greene, C. Grace Whiting, Susan Reinhard, Lynn Friss Feinberg,
Rita Choula, Jordan Green, and Ari Houser. 2015. Caregiving in the U.S. 2015.
Washington, DC: National Alliance for Caregiving & AARP Public Policy Institute.
Hunt, Gail Gibson, C. Grace Whiting, Matthew Baumgart, Lisa Weber-Raley, and Michael
Panek. 2017. Dementia Caregiving in the U.S. Washington, DC: National Alliance for
Caregiving.
Johnson, Rebecca, Kristin Harkins, Mark Cary, Pamela Sankar, and Jason Karlawish. 2015.
"The Relative Contributions of Disease Label and Disease Prognosis to Alzheimer's
Stigma: A Vignette-Based Experiment." Social Science & Medicine 143:117-127.
Jordan, Ana. 2020. "Masculinizing Care? Gender, Ethics of Care, and Fathers’ Rights
Groups." Men and Masculinities 23(1):20-41.
Kane, Robert L., and Rosalie A. Kane. 1980. "Long-Term Care: Can our Society Meet the
Needs of its Elderly?" Annual Review of Public Health:227-53.
Keefe, Bronwyn. 2017. "Pride and Prejudice: Breaking Down Socially Constructed Attitudes
and Moving Towards a ‘Convergence’ in Aging and Disability Studies." The Gerontologist
57(6):1192-1193.
Kelley, Dannielle E., Megan A. Lewis, and Brian G. Southwell. 2017. "Perceived Support
from a Caregiver's Social Ties Predicts Subsequent Care-Recipient Health." Preventive
Medicine Reports (8):108-111.
Kim, Heejung, Sangeun Lee, Jooyoung Cheon, Soyun Hong, and Mido Chang. 2018. "A
Comparative Study to Identify Factors of Caregiver Burden between Baby Boomers
and Post Baby Boomers: A Secondary Analysis of a US Online Caregiver Survey." BMC
Public Health 18:579.
Koehly, Laura M. 2017. "It's Interpersonal: Family Relationships, Genetic Risk, and
Caregiving." The Gerontologist 57(1):32-39.
Kwak, Jung, and Larry J. Polivka. 2014. "The Future of Long-Term Care and the Aging
Network." Generations: Journal of the American Society on Aging 38(2):67-73.
Lock, Margaret. 2002. Twice Dead: Organ Transplants and the Reinvention of Death. Berkeley:
University of California Press.
Lovell, Brian, and Mark A. Wetherell. 2011. "The Cost of Caregiving: Endocrine and
Immune Implications in Elderly and Non Elderly Caregivers." Neuroscience and
Biobehavioral Reviews 35(6):1342-1352.
Marwit, Samuel J., and Thomas M. Meuser. 2002. "Development and Initial Validation of an
Inventory to Assess Grief in Caregivers of Persons with Alzheimer's Disease." The
Gerontologist 42(6):751-765.
Mol, Annemarie. 2002. The Body Multiple: Ontology in Medical Practice. Durham: Duke
University Press.

95
Montgomery, Rhonda J. V., and Karl D. Kosloski. 2013. "Pathways to a Caregiver Identity
and Implications for Support Services." Pp. 131-156 in Caregiving across the Lifespan,
edited by R.C. Talley, and R.J.V. Montgomery. New York: Springer.
Moon, Heehyul, and Peggye Dilworth-Anderson. 2015. "Baby Boomer Caregiver and
Dementia Caregiving: Findings from the National Study of Caregiving." Age and Ageing
44:300-306.
Moradi, Bonnie, and Patrick R. Grzanka. 2017. "Using Intersectionality Responsibly:
Toward Critical Epistemology, Structural Analysis, and Social Justice Activism." Journal
of Counseling Psychology 64(5):500-513.
Morgan, Debra, Anthea Innes, and Julie Kosteniuk. 2011. "Dementia Care in Rural and
Remote Settings: A Systematic Review of Formal or Paid Care." Maturitas (68):17-33.
Musich, Shirley, Shaohung S. Wang, Sandra Kraemer, Kevin Hawkins, and Ellen Wicker.
2017. "Caregivers for Older Adults: Prevalence, Characteristics, and Health Care
Utilization and Expenditures." Geriatric Nursing 38:9-16.
NAC (National Alliance for Caregiving), and AARP Public Policy Institute. 2015.
"Caregiving in the U.S.", Accessed September 6, 2019.
https://www.caregiving.org/research/open-data/
National Institute on Aging, National Institutes of Health, U.S. Department of Health and
Human Services, and World Health Organization. 2011. Global Health and Aging.
Washington, DC: National Institutes of Health.
Nichols, Linda O., Jennifer Martindale-Adams, Robert Burns, Marshall J. Graney, and
Jeffrey Zuber. 2011. "Typical and Atypical Dementia Family Caregivers: Systematic and
Objective Comparisons." The International Journal of Aging and Human Development
72(1):27-43.
Oliker, Stacey. 2011. "Sociology and Studies of Gender, Caregiving, and Inequality." Sociology
Compass 5(11):968-983.
Parker, Michael W., Jeffrey M. Bellis, Phillip Bishop, Mary Harper, Richard M. Allman,
Cynthia Moore, and Paula Thompson. 2002. "A Multidisciplinary Model of Health
Promotion Incorporating Spirituality into a Successful Aging Intervention with African
American and White Elderly Groups." The Gerontologist 42(3):406-415.
Pew Research Center. 2015. The Whys and Hows of Generations Research. Washington, DC: Pew
Research Center. Accessed 2019 August 8. http://www.people-
press.org/2015/09/03/the-whys-and-hows-of-generations-research/
Piercy, Kathleen W. 2007. "Characteristics of Strong Commitments to Intergenerational
Family Care of Older Adults." The Journals of Gerontology. Series B, Psychological Sciences and
Social Sciences 62(6):S381-S387.
Pillemer, Karl, and J. J. Suitor. 2006. "Making Choices: A within-Family Study of Caregiver
Selection." The Gerontologist 46(4):439-448.
96
Pinquart, Martin, and S. Sörensen. 2003. "Differences between Caregivers and
Noncaregivers in Psychological Health and Physical Health." Psychology and Aging
18(2):250-267.
------. 2004. "Associations of Caregiver Stressors and Uplifts with Subjective Well-being and
Depressive Mood: A Meta-Analytic Comparison." Aging & Mental Health 8(5):438-449.
Raschick, Michael, and Berit Ingersoll-Dayton. 2004. "The Costs and Rewards of Caregiving
among Aging Spouses and Adult Children." Family Relations 53(3):317-325.
Redfoot, Donald, Lynn Feinberg, and Ari Houser. 2013. The Aging of the Baby Boom and the
Growing Care Gap: A Look at Future Declines in the Availability of Family Caregivers.
Washington, DC: AARP Public Policy Institute.
Remler, Dahlia K., and Gregg G. van Ryzin. 2011. Research Methods in Practice: Strategies for
Description and Causation. Thousand Oaks, CA: Sage.
Richardson, Liana J., and Tyson H. Brown. 2016. "(En)Gendering Racial Disparities in
Health Trajectories: A Life Course and Intersectional Analysis." SSM - Population Health
2:425-435.
Richman, Laura S., and Alyssa N. Zucker. 2019. "Quantifying Intersectionality: An
Important Advancement for Health Inequality Research." Social Science & Medicine
226:246-248.
Rodriguez, Javier M. 2018. "Health Disparities, Politics, and the Maintenance of the Status
Quo: A New Theory of Inequality." Social Science & Medicine 200:36-43.
Roth, David L., William E. Haley, Martha Hovater, Martinique Perkins, Virginia G. Wadley,
and Suzanne Judd. 2013. "Family Caregiving and all-Cause Mortality: Findings from a
Population-Based Propensity-Matched Analysis." American Journal of Epidemiology
178(10):1571-1578.
Roth, David L., Lisa Fredman, and William E. Haley. 2015. "Informal Caregiving and its
Impact on Health: A Reappraisal from Population-Based Studies." The Gerontologist
55(2):309-319.
Ryan, Lindsay H., Jacqui Smith, Toni C. Antonucci, and James S. Jackson. 2012. "Cohort
Differences in the Availability of Informal Caregivers: Are the Boomers at Risk?" The
Gerontologist 52(2):177-188.
Savundranayagam, Marie Y., and Rhonda J. V. Montgomery. 2010. "Impact of Role
Discrepancies on Caregiver Burden among Spouses." Research on Aging 32(2):175-199.
Scales, Kezia, Arielle Altman, Stephen Campbell, Allison Cook, Angelina Del Rio Drake,
Robert Espinoza, and Jodi M. Sturgeon. 2020. It's Time to Care: A Detailed Profile of
America's Direct Care Workforce. Bronx: PHI.
Scales, Kezia, Jodi M. Sturgeon, Robert Espinoza, Stephen Campbell, and Allison Cook.
2019. Envisioning the Future of Home Care: Trends and Opportunities in Workforce Policy and
Practice. Bronx: PHI.
97
Schulz, Richard, and Jill Eden, a Committee on Family Caregiving for Older Adults, Board
on Health Care Services, Health and Medicine Division, and National Academies of
Sciences, Engineering, and Medicine. 2016. Families Caring for an Aging America.
Washington, DC: National Academies Press.
Scommegna, Paola. 2016. "Family Caregiving." Today's Research on Aging (33). Washington,
DC: Population Reference Bureau.
Scott, Nicholas A., and Janet Siltanen. 2017. "Intersectionality and Quantitative Methods:
Assessing Regression from a Feminist Perspective." International Journal of Social Research
Methodology 20(4):373-385.
Seaman, Aaron T. 2018. "The Consequence of "Doing Nothing": Family Caregiving for
Alzheimer's Disease as Non-Action in the U.S." Social Science & Medicine 197:63-70.
Seeher, Katrin, Lee-Fay Low, Simone Reppermund, and Henry Brodaty. 2013. "Predictors
and Outcomes for Caregivers of People with Mild Cognitive Impairment: A Systematic
Literature Review." Alzheimer's & Dementia: The Journal of the Alzheimer's Association 9:346-
355.
Seltzer, Marsha M., and Lydia W. Li. 1996. "The Transitions of Caregiving: Subjective and
Objective Definitions." The Gerontologist 36(5):614-626.
------. 2000. "The Dynamics of Caregiving: Transitions during a Three-Year Prospective
Study." The Gerontologist 40(2):165-178.
Shurgot, Gia R., and Bob G. Knight. 2005. "Influence of Neuroticism, Ethnicity, Familism,
and Social Support on Perceived Burden in Dementia Caregivers: Pilot Test of the
Transactional Stress and Social Support Model." The Journals of Gerontology: Psychological
Sciences 60(6):331.
Spector, Aimee, Martin Orrell, Astrid Schepers, and Niamh Shanahan. 2012. "A Systematic
Review of ‘Knowledge of Dementia’ Outcome Measures." Ageing Research Reviews
11(1):67-77.
Stites, Shana D., Jonathan D. Rubright, and Jason Karlawish. 2018. "What Features of
Stigma Do the Public Most Commonly Attribute to Alzheimer's Disease Dementia?
Results of a Survey of the U.S. General public." Alzheimer's & Dementia: The Journal of the
Alzheimer's Association 14(7):925-932.
Suitor, J. Jill, Megan Gilligan, and Karl Pillemer. 2016. "Stability, Change, and Complexity in
Later-Life Families." in Handbook of Aging and the Social Sciences, edited by L. George and
K. Ferraro. Oxford: Elsevier Science & Technology.
Unson, Christine, Deborah Flynn, Mary A. Glendon, Elayne Haymes, and Diane Sancho.
2015. "Dementia and Caregiver Stress: An Application of the Reconceptualized
Uncertainty in Illness Theory." Issues in Mental Health Nursing 36(6):439-446.

98
Van Durme, Thérèse, Jean Macq, Caroline Jeanmart, and Micheline Gobert. 2012. "Tools
for Measuring the Impact of Informal Caregiving of the Elderly: A Literature Review."
International Journal of Nursing Studies 49(4):490-504.
Vitaliano, Peter P., Michael Murphy, Heather M. Young, Diana Echeverria, and Soo Borson.
2011. "Does Caring for a Spouse with Dementia Promote Cognitive Decline? A
Hypothesis and Proposed Mechanisms." Journal of the American Geriatrics Society
59(5):900-908.
Wang, Xiao R., Karen M. Robinson, and Lisa Carter-Harris. 2014. "Prevalence of Chronic
Illnesses and Characteristics of Chronically Ill Informal Caregivers of Persons with
Dementia." Age and Ageing 43:137-141.
Warner, David F., and Tyson H. Brown. 2011. "Understanding how Race/Ethnicity and
Gender Define Age-Trajectories of Disability: An Intersectionality Approach." Social
Science & Medicine 72(8):1236-1248.
Werner, Perla, Orit Karnieli-Miller, and Shmuel Eidelman. 2013. "Current Knowledge and
Future Directions about the Disclosure of Dementia: A Systematic Review of the First
Decade of the 21st Century." Alzheimer’s & Dementia 9:e74-e88.
West, Loraine A., Samantha Cole, Daniel Goodkind, and Wan He. 2014. 65+ in the United
States: 2010. Washington, DC: U.S. Census Bureau.
Whitlatch, Carol J., and Silvia Orsulic-Jeras. 2018. "Meeting the Informational, Educational,
and Psychosocial Support Needs of Persons Living with Dementia and their Family
Caregivers." The Gerontologist 58:58.
Willert, Bria, and Krista L. Minnotte. 2019. "Informal Caregiving and Strains: Exploring the
Impacts of Gender, Race, and Income." Applied Research in Quality of Life (2019).
Accessed May 16, 2020. doi.org/10.1007/s11482-019-09786-1
Wright, Erik Olin 2005. Approaches to Class Analysis. Cambridge: Cambridge University Press.
Zhang, Jingjun, and Diana T. F. Lee. 2017. "Meaning in Stroke Family Caregiving: A
Literature Review." Geriatric Nursing 38(1):48-56.

99