INTERSECTIONAL ANALYSIS OF VACCINATION INEQUITY AMONG INDIVIDUALS WITH DISABILITIES IN THE UNITED STATES DURING COVID-19 PANDEMIC

INTERSECTIONAL ANALYSIS OF VACCINATION INEQUITY AMONG

INDIVIDUALS WITH DISABILITIES IN THE UNITED STATES DURING

COVID-19 PANDEMIC

Chapter 1: Introduction to the Study

Introduction

In this study, I examined the intersectional impacts of disability status,

sociodemographic indicators (e.g., race/ethnicity, socioeconomic status), and geographic

location on COVID-19 outcomes (infection rates, hospitalizations, mortality) and access

to healthcare services (testing, treatment, vaccination) for individuals with different types

of disabilities in the United States during the COVID-19 pandemic. Individuals with

disabilities face unique challenges and barriers in accessing healthcare services and public

health interventions despite being at increased risk for severe illness and adverse

outcomes.

Previous researchers investigated the impact of COVID-19 on individuals with

disabilities. Still, there was a gap in comprehensive studies that analyzed the

intersectional effects of disability, sociodemographic indicators, and geographic location

on COVID-19 outcomes and healthcare access. I addressed this gap by conducting a

quantitative analysis that incorporated the intersectionality framework to provide a more

nuanced and holistic understanding of the inequities experienced by individuals with

disabilities during the pandemic.

The study's findings contributed to developing targeted interventions, policies, and

healthcare practices to mitigate inequities and improve health outcomes for individuals

with disabilities, particularly those at the intersection of multiple forms of disadvantage.

By highlighting the unique challenges faced by individuals with disabilities and

marginalized communities, the study promoted social justice and the rights of

persons with disabilities. The emphasis on intersectionality underscored the importance of

considering multiple dimensions of identity and disadvantage when addressing health

inequities, challenging existing paradigms, and promoting more inclusive and holistic

approaches to healthcare and public health policy. The findings from this study could be

used to inform evidence-based interventions, shape policy decisions, and drive positive

social change by advancing our understanding of the complex interplay between

disability, sociodemographic indicators, geographic location, and COVID-19 outcomes.

In the introduction, I offer a comprehensive study overview, including its

background, problem statement, purpose, research questions, hypotheses, theoretical

framework, nature, and significance. In the literature review, I delve into existing

research, covering topics such as the impact of COVID-19 on individuals with

disabilities, healthcare access inequities, intersectionality, and social determinants of

health. In the research design and methodology section, I outline the quantitative

crosssectional study design and detailed data sources, sampling techniques, data

collection procedures, and analysis methods.

The results section includes the findings derived from data analysis, including

descriptive and inferential statistics, and the outcomes of hypothesis testing. In the

discussion section, I interpret and contextualize the findings related to existing literature,

theoretical frameworks, and research questions. In implications section, I explore the

potential effects of the study on positive social change, policy recommendations, and

future research avenues. Finally, the conclusion includes a summary of the main findings,

acknowledged limitations, and highlighted the overall significance of the study.

Background

Research consistently showed that individuals with disabilities faced significant

inequities and barriers in accessing healthcare services and public health interventions

during the COVID-19 pandemic (Goyal et al., 2023; McBride-Henry et al., 2023).

Despite being at an increased risk for severe illness and adverse outcomes from

COVID19, this population encountered unique challenges that hindered their ability to

receive essential medical care, testing, and vaccinations.

Several studies highlighted the lower COVID-19 vaccination rates among

individuals with disabilities compared to those without disabilities (Hollis et al., 2023;

Myers et al., 2022). These inequities were attributed to vaccine hesitancy, concerns about

side effects, distrust in government information, and accessibility issues (Burdick &

Christopher, 2022; Myers et al., 2022). Additionally, research showed that individuals

with disabilities experienced higher rates of COVID-19-related hospitalizations and

mortality (Nab et al., 2023; Sosenko et al., 2023).

However, a significant gap in the existing literature was the lack of comprehensive

quantitative analyses examining the intersectional impacts of disability status,

sociodemographic indicators (such as race/ethnicity and socioeconomic status), and

geographic location on COVID-19 outcomes and access to healthcare services for

individuals with different types of disabilities. While studies explored individual aspects

of this issue, there was a need for research that integrated an intersectionality framework

to investigate the compounding effects of multiple marginalized identities on the

experiences of individuals with disabilities during the pandemic.

Addressing this gap was crucial because individuals with disabilities often faced

multiple and overlapping forms of disadvantage and marginalization, which exacerbated

the inequities they experienced in healthcare access and outcomes. By adopting an

intersectional approach, I developed a more nuanced and holistic understanding of the

complex interplay between disability, sociodemographic indicators, geographic location,

and COVID-19-related inequities.

This study was needed to inform the development of targeted interventions,

policies, and healthcare practices that could effectively address the unique needs and

vulnerabilities of individuals with disabilities, particularly those at the intersection of

multiple marginalized identities. By shedding light on the intersectional nature of these

inequities, the study could contribute to the broader efforts of promoting equity,

inclusivity, and social justice in public health responses to pandemics and other health

crises.

Problem Statement

The specific research problem that I addressed was the lack of comprehensive

quantitative analysis examining the intersectional impacts of disability status,

sociodemographic indicators (race/ethnicity, age), and vaccination status for individuals

with different types of disabilities in the United States during the COVID-19 pandemic.

There was a consensus within the research community that this problem was

current, relevant, and significant to public health, epidemiology, disability studies, and

health inequities research. The COVID-19 pandemic exposed and exacerbated existing

inequities in healthcare access and outcomes for marginalized communities, including

individuals with disabilities (Friedman & VanPuymbrouck, 2023; Turcheti et al., 2022).

Recent studies consistently demonstrated that individuals with disabilities faced

disproportionate challenges in accessing COVID-19 testing, treatment, and vaccination

services, contributing to higher rates of infection, hospitalization, and mortality (Nab et

al., 2023; Sosenko et al., 2023).

Previous researchers explored various aspects of this problem, such as the impact

of disability on COVID-19 outcomes (Peeters et al., 2023; Salmerón Ríos et al., 2021)

and the barriers to vaccination uptake among individuals with disabilities (Burdick &

Christopher, 2022; Myers et al., 2022), but there was a lack of comprehensive studies that

integrated an intersectional approach to examine the compounding effects of disability,

sociodemographic indicators, and geographic location.

In this study, I built upon and countered the limitations of previous research by

adopting an intersectional framework to investigate the complex interplay between

disability status, sociodemographic indicators (such as race/ethnicity and socioeconomic

status), and geographic location in shaping COVID-19 outcomes and access to healthcare

services for individuals with different types of disabilities. By incorporating

intersectionality, I recognized that the experiences of individuals with disabilities were

not uniform but were influenced by the intersection of various forms of marginalization

and disadvantage (Brown & Ciciurkaite, 2023; Crenshaw, 1989).

I addressed a meaningful gap in the current research literature by providing a

comprehensive and nuanced understanding of the inequities experienced by individuals

with disabilities during the COVID-19 pandemic. By quantitatively analyzing the

intersectional impacts of disability status, sociodemographic indicators, and geographic

location, I developed information that could be used to improve targeted interventions,

policies, and healthcare practices that accounted for the diverse needs and vulnerabilities

of this population, ultimately contributing to the broader efforts of promoting equity and

inclusivity in public health responses.

Purpose

The purpose of this quantitative cross-sectional study was to examine the

intersectional impacts of disability status, sociodemographic indicators (race/ethnicity,

age), and vaccination status for individuals with different types of disabilities in the

United States during the COVID-19 pandemic. By accounting for the intersection of

disability with other social determinants, I provided a more nuanced and holistic

understanding of the inequities experienced by individuals with disabilities during the

pandemic.

Variables such as socioeconomic status (income, education level), geographic

location (e.g., urban/rural, ZIP code, county), and COVID-19 outcomes (e.g., infection

rates, hospitalizations, mortality) were vital in this analysis. Still, they were unavailable in

the Household Pulse Survey (HPS) public use files. I contacted the Surveillance and

Epidemiology Branch via [email protected] and requested access to these variables but

was only granted access to the variables available in the public use files.

Research Questions and Hypotheses

Research Question 1 (RQ1): Is there an association between disability status and

COVID-19 vaccination uptake among adults aged 18 and older in the United States, and does

this association vary based on sociodemographic indicators such as race/ethnicity and age?

Null Hypothesis (H01): There is no association between disability status and

COVID-19 vaccination uptake among adults aged 18 and older in the United States,

considering the intersection with sociodemographic indicators such as race/ethnicity and

age.

Alternate Hypothesis (H11): There is an association between disability status and

COVID-19 vaccination uptake among adults aged 18 and older in the United States,

considering the intersection with sociodemographic indicators such as race/ethnicity and

age.

Research Question 2 (RQ2): Is there a difference in COVID-19 vaccination uptake

among adults with different disability types, and is this difference moderated by

sociodemographic indicators such as race/ethnicity and age?

Null Hypothesis (H02): There is no difference in COVID-19 vaccination uptake

among adults with different disability types, controlling for sociodemographic indicators

such as race/ethnicity and age.

Alternate Hypothesis (H12): There is a difference in COVID-19 vaccination

uptake among adults with different disability types, controlling for sociodemographic

indicators such as race/ethnicity and age.

Research Question 3 (RQ3): Is there an interaction effect between disability status

and reported reasons for not receiving vaccinations (COVID-19 vaccine hesitancy)

among adults aged 18 and older in the United States, and is this interaction influenced by

sociodemographic indicators such as race/ethnicity and age?

Null Hypothesis (H03): There is no interaction effect between disability status and

reported reasons for not receiving vaccinations among adults aged 18 and older in the

United States, controlling for sociodemographic indicators such as race/ethnicity and age.

Alternate Hypothesis (H13): There is an interaction effect between disability status

and reported reasons for not receiving vaccinations among adults aged 18 and older in the

United States, controlling for sociodemographic indicators such as race/ethnicity and age.

The independent variables were disability status and disability types. The

dependent variables were COVID-19 vaccination uptake (vaccination status) and reported

reasons for not receiving vaccinations. The associations being tested were the

relationships between disability status, disability types, sociodemographic indicators,

vaccination status on COVID-19 vaccination uptake, and reported reasons for not

receiving vaccinations.

The variables were measured through self-reported data from survey responses,

with disability status, disability types, sociodemographic indicators, and COVID-19

vaccination uptake (vaccination status) being a dichotomous variable (vaccinated or not

vaccinated).

Theoretical and Conceptual Framework for the Study

This study was underpinned by the concepts of intersectionality, as introduced by

Crenshaw (1989), and the social determinants of health framework proposed by the

World Health Organization (WHO). Intersectionality, initially articulated by Kimberlé

Crenshaw, acknowledged the multifaceted nature of discrimination and marginalization,

emphasizing that various factors such as race, gender, class, and disability intersected and

compounded each other, shaping individuals' experiences. In this study, I integrated

intersectionality and the social determinants of health framework to understand the

complex interplay between disability status, sociodemographic indicators, and

vaccination status among individuals with disabilities during the COVID-19 pandemic.

I used the intersectionality theory in the examination of how disability status

intersected with sociodemographic indicators and vaccination status, aiming for a

nuanced understanding of the inequities experienced by individuals with disabilities. By

adopting this lens, I uncovered how multiple marginalized identities contributed to health

inequities during the pandemic.

The social determinants of health framework complements intersectionality by

highlighting broader social, economic, and environmental factors influencing health

outcomes. This framework supported the inclusion of sociodemographic indicators and

geographic location as variables, acknowledging their role in shaping access to healthcare

services and COVID-19 outcomes for individuals with disabilities.

The study's conceptual framework was grounded in research highlighting

persistent barriers and inequities faced by individuals with disabilities during the

pandemic. This research underscored inequities in vaccination rates, infection rates,

hospitalizations, and mortality among this population. Additionally, I recognized the

compounded disadvantages resulting from the intersection of disability with factors like

race/ethnicity, socioeconomic status, and geographic location.

Logical connections among critical elements of the conceptual framework were

evident. Individuals with disabilities, especially those with cognitive or physical

disabilities, encountered challenges in accessing healthcare services and protective

measures, contributing to inequities in health outcomes. Moreover, factors like

race/ethnicity, socioeconomic status, and age intersected with disability status,

exacerbating inequities. Geographic factors further influenced healthcare access and

COVID-19 outcomes, shaping individuals' access to testing, treatment, and vaccination

services. By examining the intersectional impacts of these elements, my goal was to

understand the factors contributing to inequities among individuals with disabilities

during the pandemic, informing efforts to promote health equity and address systemic

barriers to healthcare access.

Nature of the Study

I used a quantitative cross-sectional design for this study, which was appropriate

for addressing the research questions and examining the intersectional impacts of

disability status, sociodemographic indicators (race/ethnicity, age), and vaccination status

for individuals with different types of disabilities on COVID-19 vaccination uptake and

reported reasons for not receiving vaccinations during the COVID-19 pandemic.

A quantitative approach was suitable for this study because I conducted a

systematic collection and analysis of numerical data, enabling the examination of

relationships between variables and the testing of hypotheses. The cross-sectional design

provided a snapshot of the variables of interest at a specific time, which was well-suited

for assessing the prevalence of COVID-19 outcomes and healthcare access inequities

within the target population.

The critical study variables were disability status, disability types, race/ethnicity,

age, vaccination status (COVID-19 vaccination uptake), and reported reasons for not

receiving vaccinations. I used data from the Household Pulse Survey (HPS), a national

survey conducted by the U.S. Census Bureau, to measure household experiences during

the COVID-19 pandemic. The HPS collected self-reported data on individuals' COVID19

vaccination status, disability status, sociodemographic characteristics, geographic

location, and access to healthcare services. The survey employed sampling techniques

and weighting procedures to ensure the representativeness of the target population.

I used descriptive statistics to summarize and characterize the sample and

inferential statistics, such as regression analyses (e.g., logistic regression, multiple linear

regression), to examine the relationships between the independent and dependent

variables. I employed statistical techniques such as interaction effects, stratified analyses,

and intersectional regression models to examine the compounding effects of disability

status, disability types, and sociodemographic indicators on the outcomes of interest,

addressing the intersectional nature of the research questions.

The quantitative cross-sectional design, a national survey dataset, and appropriate

statistical analyses provided valuable insights into the intersectional impacts of disability

status, disability types, and sociodemographic indicators on COVID-19 uptake and

reported reasons for not receiving vaccination for individuals with disabilities during the

pandemic. This approach aligned with my objectives, and I used it to develop a

comprehensive understanding of the inequities experienced by this vulnerable population.

Definitions

Intersectionality: Crenshaw (1989) introduced the recognition that the interaction

of multiple, intersecting social identities and systems of privilege and oppression shaped

individuals' experiences.

Sociodemographic indicators: In this study, sociodemographic characteristics

included race/ethnicity, age, and socioeconomic status (SES). Age indicated the age of the

respondent at the time of the survey.

Socioeconomic status (SES): In this study, SES was a composite measure of an

individual's economic and social position relative to others, encompassing income,

education, and occupation (APA, 2022). This definition aligned with the conventional

understanding of SES in public health research, which acknowledged the influence of

economic and social factors on health outcomes.

Disability: Disability was broadly defined in this study as physical, sensory,

cognitive, mental health, and other impairments that limited daily activities or required

assistance (WHO, 2011). This inclusive definition recognized the diverse nature of

disabilities and their impact on individuals' lives, encompassing various types and degrees

of impairment.

Vaccination Uptake: Vaccination uptake refers to the proportion of individuals

who received a vaccine among the eligible population. In this study, COVID-19

vaccination uptake pertained explicitly to the percentage of individuals vaccinated against

COVID-19 among the adult population aged 18 and older in the United States (CDC,

2020). This definition focused on the uptake of COVID-19 vaccines and distinguished it

from broader measures of vaccination coverage.

Healthcare Access Inequities: Healthcare access inequities refer to unfair, unjust,

and avoidable inequalities in the availability, utilization, quality, and outcomes of

healthcare services among different populations (Haggerty et al., 2020). These inequities

were typically rooted in systemic issues such as socioeconomic status, race, ethnicity,

geography, gender, and other social determinants of health. Unlike general healthcare

inequalities, which merely describe differences, healthcare access inequities emphasize

the ethical and moral imperative to address and rectify these inequalities. This study

examined healthcare access inequities in the context of individuals with disabilities

during the COVID-19 pandemic, focusing on identifying barriers to accessing COVID-19

testing, treatment, and vaccination services.

Race/Ethnicity: Race and ethnicity refer to the racial or ethnic identity of the

individual as self-reported (OMB, 1997).

COVID-19 Pandemic: The global outbreak of the severe acute respiratory

syndrome coronavirus 2 (SARS-CoV-2) disease was declared a public health emergency

of international concern in January 2020 (WHO, 2020).

Reported Reasons for not Receiving Vaccination (Vaccine Hesitancy): Reported

reasons for not receiving vaccination in this study can be simply defined as COVID-19

vaccine hesitancy. Vaccine hesitancy among individuals with disabilities refers to

a delay in acceptance or refusal of COVID-19 vaccines by people with disabilities despite

the availability of vaccination services (Myers et al., 2022).

Assumptions

The key assumptions for this study were as follows: firstly, I assumed that the

self-reported data collected through the HPS accurately reflected individuals' disability

status, sociodemographic characteristics, COVID-19 vaccination status, and access to

healthcare services. The study relied on self-reported data, which might have been subject

to reporting biases. However, self-reported data are commonly used in public health

research, and the HPS employed measures to ensure data quality and reliability.

Validating self-reported data through external sources was not feasible within the scope of

this study.

Secondly, I assumed that the sample obtained from the HPS was representative of

the target population of adults aged 18 and older in the United States. The HPS used

sampling techniques and weighting procedures to ensure the sample was representative of

the national population. However, relying on an online survey might have introduced

biases related to digital access and literacy, potentially underrepresenting specific

subgroups within the disability community. I acknowledged this limitation, and the

findings were interpreted within the context of the study population.

The intersectionality framework accurately captured the complex interplay

between disability status, sociodemographic indicators (race/ethnicity, age), and

vaccination status for individuals with different types of disabilities on COVID-19

vaccination uptake and reported reasons for not receiving vaccinations. The

intersectionality framework was a well-established theoretical social science and public

health research perspective. I assumed that this was a suitable approach for examining the

multidimensional and intersecting forms of disadvantage experienced by individuals with

disabilities during the COVID-19 pandemic.

Scope and Delimitations

In this study, I examined the intersectional impacts of disability status,

sociodemographic indicators (race/ethnicity, age), and vaccination status for individuals

with different types of disabilities on COVID-19 vaccination uptake and reported reasons

for not receiving vaccinations during the COVID-19 pandemic. I chose this focus to

address the significant gap in the existing literature, which needed comprehensive

quantitative analyses that integrated an intersectional approach to investigate these

inequities.

The study was delimited to the adult population aged 18 and older in the United

States. I selected this population as the primary focus due to the availability of relevant

data from the HPS, which collected self-reported information on COVID-19-related

experiences and outcomes for this age group. The exclusion of individuals under 18 was a

delimitation of the study, as the experiences and needs of children and adolescents with

disabilities might have differed from those of adults.

Regarding theoretical and conceptual frameworks, the study was primarily

grounded in the intersectionality theory and the social determinants of health framework.

While other relevant theories and models (e.g., the disability rights framework and the

ecological model of health) might have provided additional insights, the study was

delimited to these critical theoretical perspectives to maintain a focused and manageable

scope.

Regarding generalizability, this study's findings primarily apply to the adult

population with disabilities residing in the United States during the COVID-19 pandemic.

Using a nationally representative dataset, such as the HPS, and applying appropriate

sampling and weighting techniques were expected to enhance the generalizability of the

results to the broader U.S. adult population with disabilities. However, the study's

crosssectional nature and the potential biases associated with self-reported data might

have limited the generalizability of the findings beyond the study period and specific

population characteristics. Cautious interpretation and acknowledgment of these

limitations are necessary when discussing the broader implications of the study's results.

Overall, the scope and delimitations of this study were designed to provide a

comprehensive and focused examination of the intersectional inequities experienced by

individuals with disabilities in the United States during the COVID-19 pandemic while

acknowledging the limitations in terms of population, theoretical frameworks, and

generalizability.

Limitations

One fundamental limitation affecting the generalizability of the study was its

reliance on self-reported data from the Household Pulse Survey (HPS). While the HPS

was a national survey, the inherent biases of self-reported data, such as social desirability

and recall bias, could have impacted the accuracy and reliability of the findings.

Additionally, the study focused on adults aged 18 and older in the United States,

excluding individuals under 18. This exclusion limited the applicability of the results to

the broader population, particularly children and adolescents with disabilities who might

have experienced different healthcare challenges and outcomes.

The trustworthiness of the study's findings was constrained by the potential biases

associated with the online survey format of the HPS. Digital access and literacy issues

might have led to the underrepresentation of specific subgroups within the disability

community, such as those with limited internet access or lower digital literacy. This

underrepresentation could have skewed the results, making them less reflective of the

disabled population. Furthermore, the study's cross-sectional design only provided a

snapshot of the data at a specific point in time, limiting the ability to draw causal

inferences or observe changes over time.

The study's internal validity was influenced by the quality of the self-reported

data, which may not have always accurately reflected individuals' actual disability status,

sociodemographic characteristics, and healthcare experiences. Reporting biases, such as

over- or under-reporting of vaccination status and healthcare access issues, could have

affected the validity of the findings. Additionally, the study could not validate the

selfreported data against external sources, further impacting its internal validity. The

complexity of measuring intersectionality through logistic regression models might have

also introduced challenges in accurately capturing the nuanced interplay of multiple

social identities and their compounded effects on health outcomes.

Reliability issues arose from the study's reliance on a single HPS dataset, which

may not have consistently captured all relevant variables over time. The dynamic nature

of the COVID-19 pandemic and changing public health policies could have led to

variations in survey responses, impacting the consistency and repeatability of the

findings. Additionally, self-reported measures for critical variables, such as vaccination

uptake and reasons for vaccine hesitancy, might have been subject to individual

perceptions and reporting accuracy fluctuations, further affecting the study's reliability.

Significance

This study contributed to the growing body of research on the impact of the

COVID-19 pandemic on individuals with disabilities by adopting an intersectional

approach. By examining the interplay between of disability status, disability types, and

sociodemographic indicators on COVID-19 uptake and reported reasons for not receiving

vaccination for individuals with disabilities during the pandemic, the study provided a

more nuanced understanding of the multidimensional inequities experienced by this

population during the pandemic. The findings expanded the academic knowledge in

public health, epidemiology, disability studies, and health inequities research.

Moreover, the study's focus on COVID-19 vaccination uptake and the reported

reasons for not receiving vaccinations among individuals with different disability types

contributed to the limited research in this area. By elucidating the factors that influenced

vaccination decisions and barriers to access, the study informed a more comprehensive

understanding of the unique challenges faced by individuals with disabilities in accessing

this vital public health intervention.

Furthermore, this study's integration of the intersectionality framework and the

social determinants of health approach advanced the application of these theoretical

concepts in health-related research. By demonstrating the value of an intersectional lens

in examining health inequities, the study encouraged further research that considered the

complex interplay of multiple social identities and determinants of health.

Additionally, the findings of this study informed the development of targeted

public health interventions, healthcare practices, and policies aimed at addressing the

specific needs and barriers faced by individuals with disabilities during public health

emergencies. The insights gained on the intersectional factors influencing COVID-19

vaccination uptake and healthcare access guided the design of tailored outreach,

education, and service delivery strategies for this population.

Besides, the study's emphasis on the unique challenges and inequities experienced

by individuals with disabilities contributed to developing more inclusive and equitable

pandemic preparedness and response plans. The evidence generated informed integration

of the disability community's perspectives and needs into public health emergency

planning and decision-making processes.

Also, by highlighting the intersectional barriers and inequities faced by

individuals with disabilities, this study contributed to advocacy efforts to promote equity

and inclusion in healthcare and public health services. The study's findings informed

initiatives that challenged existing paradigms and promoted more holistic, personcentered

approaches to addressing the needs of individuals with disabilities, especially during

public health crises.

Finally, the study's focus on the intersection of disability, sociodemographic

indicators, and COVID-19 outcomes aligned with the principles of the United Nations

Convention on the Rights of Persons with Disabilities, which emphasized the right to the

highest attainable standard of health and equal access to healthcare services. By

uncovering the unique challenges and barriers faced by individuals with disabilities, this

research contributed to the advocacy and policymaking efforts aimed at upholding the

rights and improving the overall well-being of persons with disabilities.

Summary

This chapter introduced the study's aim to examine the intersectional impacts of

disability status, sociodemographic indicators (race/ethnicity, age), and vaccination status

on COVID-19 vaccination uptake and reported reasons for not receiving vaccinations

among individuals with different types of disabilities during the COVID-19 pandemic. It

outlined the background, problem statement, purpose, research questions, hypotheses,

theoretical and conceptual frameworks, study nature, definitions, assumptions, scope and

delimitations, limitations, and significance of the research.

The concepts of intersectionality and the social determinants of health framed this

study. These provided a solid theoretical foundation for understanding the complex

interplay between disability, sociodemographic indicators, and health outcomes. The three

research questions investigated the associations between these variables and how they

may contribute to inequities in vaccination uptake and access to healthcare services.

The proposed research design employed a quantitative cross-sectional approach,

utilizing secondary data from the Household Pulse Survey (HPS) to analyze the relevant

variables. The data analysis plan included descriptive statistics, multivariate regression

analyses, intersectional modeling, and sensitivity analyses to address the research

questions and hypotheses. The study acknowledged several limitations, such as the

reliance on self-reported data, the cross-sectional design, and potential biases. However,

the researchers outlined reasonable measures that addressed these limitations and

enhanced the findings' validity, reliability, and generalizability.

The significance of the study lies in its potential to advance knowledge in public

health, epidemiology, disability studies, and health inequities research. By providing a

comprehensive understanding of the intersectional impacts of disability,

sociodemographic indicators, and vaccination status on COVID-19 outcomes, the study

aimed to inform targeted interventions, shape equitable policies, and promote positive

social change that upheld the rights and well-being of individuals with disabilities,

especially during public health emergencies.

Having established the research design alignment and outlined the significance of

the proposed study, the next chapter delved deeper into the review of the existing

literature. Chapter 2 provided a comprehensive synthesis of the relevant research on the

impact of the COVID-19 pandemic on individuals with disabilities, focusing on the

intersections of disability, sociodemographic indicators, and healthcare access. This

indepth literature review further contextualized the research problem, identified gaps in

the current knowledge, and solidified the rationale for the current study.

Chapter 2: Literature Review

Introduction

The specific research problem that I addressed in this study was the lack of

comprehensive quantitative analysis examining the intersectional impacts of disability

status, sociodemographic indicators (race/ethnicity, age), and vaccination status for

individuals with different types of disabilities on COVID-19 vaccination uptake and

reported reasons for not receiving vaccinations during the COVID-19 pandemic.

The purpose of this quantitative cross-sectional study was to examine the

intersectional impacts of disability status, sociodemographic indicators (race/ethnicity,

age), and vaccination status for individuals with different types of disabilities on

COVID19 vaccination uptake and reported reasons for not receiving vaccinations during

the COVID-19 pandemic. By accounting for the intersection of disability with other

social determinants, I provided a more nuanced and holistic understanding of the

inequities experienced by individuals with disabilities during the pandemic.

The existing literature highlighted the disproportionate impact of the COVID-19

pandemic on individuals with disabilities, who faced significant challenges in accessing

essential healthcare services, including testing, treatment, and vaccination. Studies

demonstrated that individuals with disabilities experienced higher rates of COVID-19

infection, hospitalization, and mortality compared to the general population (Hollis et al.,

2023; Nab et al., 2023).

Furthermore, the literature documented the persistent barriers and inequities faced

by individuals with disabilities in accessing COVID-19 vaccines, with lower vaccination

rates observed in this population (Burdick & Christopher, 2022; Myers et al., 2022).

These inequities were exacerbated by factors such as disability type, sociodemographic

characteristics, and the intersection of multiple marginalized identities (Dekker et al.,

2022; Wiggins et al., 2022).

However, the current body of research lacked a comprehensive, intersectional

analysis with an examination of the combined influence of disability status,

sociodemographic indicators (race/ethnicity, age), and vaccination status on COVID-19

outcomes and healthcare access for individuals with different types of disabilities.

This gap in the literature underscored the need for this study to address this critical

research problem.

In this chapter, I examine the literature pertinent to this study's research problem

and objectives. The chapter includes several vital sections. Firstly, in the section on

Disability and the COVID-19 Pandemic, I synthesize current evidence concerning the

disproportionate impact of the COVID-19 pandemic on individuals with disabilities,

encompassing heightened rates of infection, hospitalizations, and mortality within this

demographic. Secondly, in Barriers to Healthcare Access for Individuals with Disabilities,

I delve into the unique hurdles faced by individuals with disabilities in accessing vital

healthcare services during the pandemic, including COVID-19 testing, treatment, and

vaccination.

Thirdly, in the section Intersectionality and Social Determinants of Health, I

explore theoretical frameworks such as intersectionality and social determinants of health,

elucidating their relevance in understanding the intricate interplay between disability,

sociodemographic indicators, and health outcomes. Next, in Inequities in

COVID-19 Vaccination Uptake, I review existing literature on the inequities observed in

COVID-19 vaccination rates among individuals with disabilities, alongside factors

contributing to these inequities.

Lastly, in the section on Gaps in the Literature, I identify and critically evaluate

gaps within current research, emphasizing the necessity for the proposed intersectional

study to bridge these gaps and provide a comprehensive understanding of the

multifaceted factors influencing COVID-19 outcomes and healthcare access for

individuals with disabilities. I used my literature review as the groundwork for my study,

substantiating the significance and urgency of the research problem and the imperative

for the intended investigation.

Literature Search Strategy

For this literature review, I accessed the library databases and search engines:

PubMed, CINAHL Plus, MEDLINE, APA PsycINFO, APA PsycArticles, Embase,

ProQuest Health, SocINDEX, and Cochrane Library. I used key search terms and

combinations, including: COVID-19 OR coronavirus AND disability OR disabled OR

disabilities, COVID-19 AND vaccination AND disability OR disabilities, COVID-19

AND healthcare access AND disability OR disabilities, COVID-19 AND health inequities

AND disability OR disabilities, COVID-19 AND intersectionality AND disability OR

disabilities, COVID-19 AND social determinants of health AND disability

OR disabilities.

I reviewed literature from 2020 to the present, with a particular emphasis on

recent and pertinent literature emerging during the COVID-19 pandemic. I considered

various types of literature, including peer-reviewed journal articles, conference

proceedings, and relevant grey literature such as government reports, policy briefs, and

white papers. Alongside the library databases mentioned earlier, I scrutinized the

reference lists of relevant articles to uncover additional sources contributing to the

discourse on the intersectional impacts of disability and COVID-19.

In cases with limited current research explicitly addressing the intersectional

impacts of disability, sociodemographic indicators, and COVID-19 outcomes, I expanded

the search strategy to include literature on the broader topics of disability, health

inequities, and the social determinants of health. This resulted in a more comprehensive

understanding of the theoretical and empirical foundations that inform the proposed

study.

Furthermore, to ensure the inclusion of seminal literature, I identified vital

publications and classic works on intersectionality, social determinants of health, and

disability studies. I incorporated them into the review, even if they did not address the

COVID-19 pandemic directly. By employing this comprehensive literature search

strategy, I gathered the most relevant and up-to-date evidence to establish the research

problem, justify the significance of the study, and identify the gaps in the existing

knowledge that I sought to address.

Theoretical Foundation

This study was framed by the concept of intersectionality, initially coined by

Kimberlé Crenshaw (1989), and the social determinants of health framework proposed by

WHO. Kimberlé Crenshaw's seminal work introduced the concept of intersectionality,

which was an illustration of the multidimensional and intersecting nature of various forms

of discrimination and marginalization, such as race, gender, class, and disability.

Crenshaw argued that the experiences of individuals with multiple marginalized identities

could not be adequately captured by examining these identities in isolation, as they

intersected and compounded each other in complex ways.

The intersectionality framework has been widely applied in various disciplines,

including disability studies, health inequity research, and social justice advocacies.

Acknowledging the intersecting nature of different social identities and power structures,

the intersectionality approach was instrumental in highlighting individuals' unique

experiences and challenges at the intersection of multiple marginalized identities.

In the context of this study, the intersectionality framework was particularly

relevant as I used it for an examination of the compounding effects of disability status,

sociodemographic indicators (race/ethnicity, age), and vaccination status for individuals

with different types of disabilities during the pandemic. I used this approach to challenge

the tendency to view these factors in isolation and produce a more comprehensive and

nuanced understanding of the inequities experienced by individuals with disabilities

during the pandemic.

As the WHO proposed, the social determinants of health framework is used to that

a wide range of social, economic, and environmental factors, including socioeconomic

status, race/ethnicity, gender, disability status, and geographic location, shaped an

individual's health. These determinants influence an individual's access to resources,

exposure to risk factors, and overall health outcomes.

The social determinants of health framework have been extensively applied in

public health research, policy, and interventions. By acknowledging the broader societal

and structural factors contributing to health inequities, this approach has been

instrumental in shifting the focus from individual-level factors to the systemic and

environmental influences on health and well-being. In the context of this study, I used the

social determinants of health framework as a complementary lens to the intersectionality

approach to consider the complex interplay between disability status, sociodemographic

indicators (race/ethnicity, age), and vaccination status for individuals with different types

of disabilities during the pandemic.

The integration of the intersectionality and social determinants of health

frameworks in this study aligned with the research questions and objectives. By adopting

these theoretical perspectives, I examined the intersectional impacts of disability status,

sociodemographic indicators (race/ethnicity, age), and vaccination status on COVID-19

vaccination uptake among individuals with different disabilities. Secondly, I investigated

how the intersection of these factors shaped the reported reasons for not receiving

COVID-19 vaccinations. Finally, I developed a more holistic understanding of the

multidimensional inequities and inequities experienced by individuals with disabilities

during the COVID-19 pandemic.

I built upon and challenged existing theory by moving beyond simplistic, single

factor analyses and embracing the complexity of the lived experiences of individuals with

disabilities. My goal was to generate new insights to inform more inclusive and

equityfocused approaches to healthcare and public health interventions by applying an

intersectional lens and the social determinants of health framework.

Conceptual Framework

The key concepts and phenomena underpinning this study were disability and the

COVID-19 pandemic, intersectionality and health inequities, and social determinants of

health. Disability is a multidimensional concept encompassing a range of physical,

sensory, cognitive, and psychosocial impairments that could interact with various barriers

to hinder an individual's full and effective participation in society (WHO, 2001). The

experience of disability is shaped by the complex interplay between an individual's health

condition, personal factors, and environmental factors.

During the COVID-19 pandemic, individuals with disabilities were

disproportionately affected, facing increased risks of infection, hospitalization, and

mortality (Nab et al., 2023; Sosenko et al., 2023). The pandemic also exacerbated this

population's barriers to healthcare access and social participation, leading to widening

inequities in health outcomes (Friedman & VanPuymbrouck, 2023; Turcheti et al., 2022).

Intersectionality, as conceptualized by Kimberlé Crenshaw (1989), states that the

intersection of multiple, overlapping social identities and systems of privilege and

oppression shaped individuals' experiences. This framework challenged the tendency to

view social identities, such as disability, race, and socioeconomic status, in isolation.

Instead, it emphasized the need to examine their complex and compounding effects on

health outcomes and access to resources. In the context of health inequities research, the

intersectionality approach was instrumental in highlighting the unique experiences and

challenges faced by individuals with multiple marginalized identities (Harari & Lee,

2021). By acknowledging the intersecting nature of these identities, researchers could

better understand the systemic barriers and inequities that contributed to health inequities.

As proposed by WHO, the social determinants of health framework posited that a

wide range of social, economic, and environmental factors shape an individual's health

and well-being (WHO, n.d.). These determinants included, but were not limited to,

socioeconomic status, education, employment, housing, access to healthcare, and

discrimination. This framework showed that health was not solely an individual

responsibility but was heavily influenced by the broader social, political, and economic

systems in which people lived. By addressing these social determinants, public health

researchers and policymakers could work to reduce health inequities and promote more

equitable health outcomes.

In this study, I integrated the conceptual frameworks of disability,

intersectionality, and social determinants of health to provide a comprehensive

understanding of the multidimensional inequities experienced by individuals with

disabilities during the COVID-19 pandemic. By examining the intersection of disability

status, sociodemographic indicators (race/ethnicity, age), and vaccination status, my goal

was to elucidate how these overlapping identities and social determinants shaped

COVID-19 vaccination uptake and healthcare access for this population. I used these

conceptual frameworks to move beyond simplistic, single factor analyses and address the

complex, systemic barriers that contributed to the disproportionate impact of the

pandemic on individuals with disabilities. Furthermore, integrating these frameworks

aligned with my objective to inform evidence-based interventions, policies, and advocacy

efforts that promoted equity, inclusion, and the well-being of individuals with disabilities,

particularly during public health emergencies like the COVID-19 pandemic.

Literature Review Related to Key Variables and Concepts

Intersectionality

The concept of intersectionality, initially introduced by Kimberlé Crenshaw in

1989, gained significant traction in social sciences and public health research.

Intersectionality states that individuals' experiences are shaped by intersecting social

identities and systems of privilege and oppression (Crenshaw, 1989). In the context of

health inequities research, intersectionality shows that various social categories, such as

race, ethnicity, gender, disability, and socioeconomic status, intersected to produce unique

health and well-being experiences.

Studies showed that individuals with intersecting marginalized identities often

experienced compounded forms of discrimination and disadvantage. For example, Breaux

and Rooks (2022) investigated the intersectional effects of race/ethnicity and disability on

flu vaccine uptake among US adults aged 18 and older. Using data from the National

Health Interview Survey, the researchers found significant interactions between

race/ethnicity and disability, influencing flu vaccine uptake across different age groups.

Another study by Marfo et al. (2024) examined the intersectional dynamics of

social privilege and disadvantage in shaping access to COVID-19 information and

vaccines among ethnically diverse parents in Canada. Through semi-structured interviews

with 48 participants, including both non-Indigenous and Indigenous individuals from

various provinces, the study revealed how historical and contemporary experiences of

racism, particularly within government and medical institutions, created barriers to trust

and access to COVID-19 resources. These findings highlighted the importance of

considering multiple social identities when addressing health inequities and developing

interventions to promote equitable access to preventive healthcare services.

Intersectionality has been applied in various research areas, including health

inequities, education, criminal justice, and workplace dynamics. In healthcare, for

instance, researchers have used intersectionality to examine how race, gender, and

socioeconomic status intersected to shape health outcomes, access to care, and healthcare

experiences (Harari & Lee, 2021). Intersectionality provided a more nuanced

understanding of social inequalities by moving beyond single-axis approaches that

focused on one dimension of identity. It highlighted the complexity of individuals' lives

and experiences and underscored the need for holistic, intersectional analyses in research

and policymaking.

One area of debate centered on operationalizing and measuring intersectionality in

research. Critics argued that intersectionality was challenging to quantify and

operationalize, making it difficult to apply in empirical studies (Harari & Lee, 2021).

Ongoing discussions about the most appropriate methodological approaches for capturing

intersectional identities and experiences existed. Some researchers raised concerns about

the potential for essentializing identities or overlooking intra-group diversity within

intersecting categories (Holman et al., 2021). For example, not all individuals within a

particular racial or gender group have identical experiences, and intersectionality should

account for this diversity.

The role of privilege within intersectionality frameworks was another point of

contention. While intersectionality often focuses on marginalized identities and

experiences, it also acknowledges that individuals may hold privileged identities that

confer advantages in specific contexts (Kelly et al., 2021). However, there needs to be

more debate about addressing privilege within intersectional analyses without detracting

from the focus on marginalized groups.

Sociodemographic Indicators

Sociodemographic indicators encompassed a range of characteristics, including

but not limited to race/ethnicity, age, gender, socioeconomic status (SES), education

level, marital status, and geographic location (Beatty Moody et al., 2021). These factors

were widely recognized as determinants of health, influencing individuals' access to

resources, exposure to risks, and health-related behaviors. Numerous studies

demonstrated associations between sociodemographic indicators and various health

outcomes. For example, individuals from lower SES backgrounds tended to experience

higher rates of chronic diseases, lower life expectancy, and poorer health outcomes

compared to those from higher SES backgrounds (Kim, 2022).

Race and ethnicity were extensively studied in health inequities, with racial and

ethnic minority groups often facing disproportionate burdens of disease, reduced access

to healthcare, and inequities in healthcare quality (Javed et al., 2022). Discrimination,

socioeconomic disadvantage, and cultural differences contributed to these inequities. Age

was another critical sociodemographic factor influencing health outcomes and healthcare

utilization patterns. Older adults often experience age-related health challenges and may

require different healthcare services than younger age groups (Allen et al., 2022).

While there was consensus on the importance of sociodemographic indicators in

shaping health outcomes, there were debates regarding the relative contributions of each

factor and the mechanisms underlying these associations (Holman et al., 2021). For

example, some studies suggested that race/ethnicity may have substantially influenced

specific health outcomes more than SES, while others emphasized the role of SES in

driving health inequities.

The intersectionality of sociodemographic indicators complicated the

interpretation of the study findings. Individuals may have held multiple marginalized

identities (e.g., being a racial minority and low SES), and the combined effects of these

intersecting factors may have amplified health inequities (Vohra-Gupta et al., 2022).

However, the extent to which intersectionality influenced health outcomes remained an

ongoing area of research and debate. There needs to be more consistency in the literature

regarding the relationship between education level and health outcomes (Raghupathi &

Raghupathi, 2020). While higher levels of education were generally associated with better

health outcomes, the strength and direction of this association may have varied across

different populations and health indicators.

Disability

Disability was broadly defined in this study to include physical, sensory,

cognitive, mental health, and other impairments that limited daily activities or required

assistance. This study acknowledged the diverse nature of disabilities and their impact on

individuals' lives, encompassing various types and degrees of impairment. Individuals

with disabilities face unique challenges in accessing healthcare services, including

COVID-19 testing, treatment, and vaccination, which may have exacerbated existing

health inequities (Clemente et al., 2022; Gréaux et al., 2023).

Numerous studies documented the significant impact of disability on various

aspects of life, including physical and mental health, social relationships, employment,

education, and access to healthcare services. Individuals with disabilities often experience

barriers to full participation in society and may face stigma, discrimination, and social

exclusion.

Disability was associated with a higher prevalence of chronic health conditions,

functional limitations, and lower quality of life than the general population (Fong, 2019).

Health inequities among individuals with disabilities were well-documented, with higher

rates of preventable diseases, unmet healthcare needs, and poorer health outcomes.

Access to healthcare services was a critical issue for individuals with disabilities, with

many facing barriers such as physical inaccessibility, lack of accommodations, inadequate

provider training, and financial constraints (Gréaux et al., 2023). These barriers

contributed to healthcare utilization and the perpetuation of health inequities.

There was debate within the literature regarding the measurement and

classification of disability. Different studies may have used varying definitions and

criteria for identifying disability, leading to inconsistent prevalence estimates and

population comparisons. Disability intersected with other sociodemographic indicators

such as race, ethnicity, gender, and socioeconomic status, complicating the interpretation

of study findings (Dorsey Holliman et al., 2023). The interaction between disability and

other social identities may have amplified or mitigated the effects of disability on health

outcomes and social participation. Some researchers argued that the medical model of

disability, which focused on individual impairments and limitations, failed to capture the

broader social and environmental factors that contributed to disability and shaped

individuals' experiences. A shift towards a social model of disability, which emphasized

the role of societal barriers and discrimination, was advocated as a more comprehensive

approach to understanding disability (Zaks, 2023).

Vaccination Uptake

Vaccination uptake refers to the proportion of individuals who have received a

vaccine among the eligible population. Research on vaccination uptake has focused on

various vaccines, including those for infectious diseases like influenza, measles, and

COVID-19. Variables influencing vaccination uptake included individual characteristics

(e.g., age, race/ethnicity, socioeconomic status), access to healthcare services, vaccine

efficacy, and safety perceptions, vaccine mandates or policies, and social and cultural

factors (Kolobova et al., 2022).

Numerous studies have consistently found inequities in vaccination uptake based

on sociodemographic indicators. For example, older adults and individuals from higher

socioeconomic backgrounds generally had higher vaccination rates than younger

individuals and marginalized communities (AlShurman et al., 2021). Historically

marginalized communities, including Black, Indigenous, and Hispanic populations, often

faced barriers such as lack of access to healthcare services, mistrust of healthcare

providers, and systemic racism, which contributed to lower vaccination rates (Roat et al.,

2022). Socioeconomic status also played a significant role in vaccination uptake.

Individuals from lower socioeconomic backgrounds may have encountered financial

barriers, limited access to healthcare facilities, and inadequate health education, which

could have impeded their ability to receive vaccinations.

There were many reasons for inequities in vaccine uptake, including

discrimination, mistrust, language and cultural barriers, etc. Discrimination experienced

by specific population groups, such as racial and ethnic minorities or individuals with

disabilities, could have contributed to mistrust of healthcare systems and vaccine

hesitancy. Historical instances of medical racism and unethical research practices have

led to enduring mistrust within these communities, impacting their willingness to receive

vaccines (Morgan et al., 2022). Language and cultural differences could also have

affected vaccination uptake. Individuals from immigrant or non-English-speaking

backgrounds may have encountered challenges in understanding vaccination information,

navigating healthcare systems, and accessing culturally competent care, leading to

inequities in vaccine uptake (Salib et al., 2022).

While various interventions have been implemented to address inequities in

vaccination uptake, such as targeted outreach programs, community engagement

initiatives, and culturally tailored interventions, their effectiveness in reducing inequities

has remained mixed. Some interventions may have had limited reach or effectiveness in

addressing underlying structural barriers (Adeagbo et al., 2022). The role of healthcare

providers in addressing vaccine-related discrimination and inequities was complex. While

healthcare providers could have been crucial in building trust and promoting vaccination

uptake, discrimination or bias within healthcare settings may have further exacerbated

inequities (Allen et al., 2022). The intersectionality of social identities, such as race,

ethnicity, gender, and disability, complicated the relationship between discrimination and

vaccination uptake. Research exploring how multiple intersecting factors contributed to

inequities in vaccination uptake was still emerging and required further investigation

(Breaux & Rooks, 2022).

Healthcare Access Inequities

Healthcare access inequities refer to unfair, unjust, and avoidable inequalities in

healthcare services' availability, utilization, quality, and outcomes among different

populations (Haggerty et al., 2020; Okonkwo et al., 2020). In this study, healthcare access

inequities were examined among individuals with disabilities during the COVID19

pandemic. This included identifying barriers to accessing COVID-19 testing, treatment,

and vaccination services and inequities in healthcare outcomes.

Numerous studies have consistently documented inequities in healthcare access

based on factors such as race/ethnicity, socioeconomic status (SES), gender, geographic

location, and disability status. Racial and ethnic minorities, including Black, Hispanic,

and Indigenous populations, often face barriers to accessing healthcare services due to

systemic racism, language barriers, discrimination, and lack of culturally competent care

(Banaji et al., 2021).

Individuals with lower SES, often measured by income, education, and

occupation, experienced poorer healthcare access than those with higher SES. Economic

factors such as lack of health insurance, transportation issues, and out-of-pocket costs

contributed to these inequities (McMaughan et al., 2020). Gender inequities in healthcare

access existed, with women sometimes facing challenges related to reproductive health

services, maternal care, and access to specialty care (Tesha et al., 2023).

While it was widely acknowledged that inequities existed, an ongoing debate

existed about the underlying causes and mechanisms driving these inequities. Some

researchers emphasized social determinants of health, such as poverty, racism, and social

exclusion, as root causes, while others focused on individual behaviors and healthcare

system factors (Yearby et al., 2022). The role of health insurance coverage in mitigating

healthcare access inequities was debated. While having health insurance was generally

associated with better access to care, inequities persisted even among insured populations,

indicating that insurance alone may not have been sufficient to address all barriers

(Crowley et al., 2020). Studies examining the intersectionality of multiple social

identities, such as race/ethnicity, gender, and SES, in healthcare access were still

relatively limited. Understanding how these intersecting factors compounded or mitigated

inequities was an area of ongoing research (Vohra-Gupta et al., 2022).

Race/Ethnicity

Race and ethnicity were complex social constructs encompassing individuals' self-

identified racial or ethnic identities (White et al., 2020). In research, race and ethnicity

served as a proxy for social, cultural, and historical factors influencing health outcomes,

healthcare access, and healthcare utilization, and understanding the role of race/ethnicity

in health inequities required examining how structural racism, discrimination,

socioeconomic status, cultural beliefs, and access to healthcare intersected to shape health

outcomes within racial/ethnic groups.

Numerous studies have documented health inequities based on race/ethnicity, with

racial/ethnic minority groups often experiencing poorer health outcomes compared to

white populations. These inequities spanned various health indicators, including mortality

rates, chronic disease prevalence, access to healthcare services, and vaccination rates

(Yaya et al., 2020). Structural racism and discrimination contributed to health inequities

by limiting opportunities for socioeconomic advancement, exacerbating poverty, and

perpetuating unequal access to healthcare resources and services

(Churchwell et al., 2020). Marginalized racial/ethnic groups faced systemic barriers that

affected their physical and mental health outcomes. Cultural beliefs, traditions, and

socioeconomic factors within racial/ethnic communities influenced health behaviors,

healthcare-seeking behaviors, and treatment preferences. Understanding these factors was

crucial for developing culturally competent healthcare interventions and addressing

inequities (Nair & Adetayo, 2019).

While health inequities based on race/ethnicity were well-documented, there was

ongoing debate about the underlying causes and mechanisms driving these inequities.

Some researchers emphasized the role of socioeconomic factors and access to healthcare,

while others highlighted the impact of systemic racism and discrimination (Yearby et al.,

2022). Classifying individuals into racial/ethnic categories could be challenging and

might not fully capture the complexities of racial and ethnic identities. The use of

selfreported race/ethnicity data in research might oversimplify individuals' identities and

fail to account for intersectional experiences.

COVID-19 Pandemic

The COVID-19 pandemic was the global outbreak of the severe acute respiratory

syndrome coronavirus 2 (SARS-CoV-2) disease. It encompassed various aspects,

including epidemiology, public health measures, healthcare systems' responses,

socioeconomic impacts, and individual behaviors (Muralidar et al., 2020). Understanding

the multifaceted nature of the pandemic was essential for addressing its challenges and

mitigating its impact on global health and society.

The pandemic had profound socioeconomic impacts, including employment,

education, supply chains, and economic stability disruptions. Vulnerable populations,

such as low-income individuals, racial/ethnic minorities, and those in precarious

employment, were disproportionately affected (Tai et al., 2021). While public health

measures such as lockdowns, mask mandates, and vaccination proved effective in curbing

transmission, there needed to be more clarity about their implementation, duration, and

societal impacts (Talic et al., 2021). Controversies existed regarding the balance between

public health objectives and individual freedoms.

Misinformation targeting disability communities and the absence of tailored

information led to misunderstandings about vaccine safety and efficacy. Social and

psychological factors, such as the influence of caregivers, family members, or community

leaders who were vaccine-hesitant, and higher levels of social isolation reducing access to

accurate information, further exacerbated vaccine hesitancy, along with psychological

stress and mental health issues.

COVID-19 Vaccine Hesitancy

Numerous studies have reported that individuals with disabilities faced numerous

specific challenges during the COVID-19 vaccination pandemic (Goyal et al., 2023).

People with disabilities encountered physical and communication barriers at vaccination

sites, such as a lack of ramps, elevators, transportation, and hearing or visual

impairments, respectively (Sebring et al., 2022). Distrust in the healthcare system was

prevalent due to past experiences of discrimination or inadequate care, historical neglect,

and fears of being deprioritized or receiving lower quality care (Powell, 2020).

Healthrelated concerns included fears of adverse reactions due to existing conditions,

worries about interactions between the vaccine and ongoing treatments, and heightened

anxiety about managing potential side effects without adequate support (Rodrigues et al.,

2022).

Individuals who belong to multiple marginalized groups, such as a Black person

with a disability living in poverty, face barriers that are not merely additive but

multiplicative, significantly intensifying their overall experience of disadvantage and

skepticism (Wickenden, 2023). Public health messages and interventions often fail to

consider the cultural and social contexts of intersecting identities, rendering them less

effective. Additionally, those with intersecting marginalized identities may be more

vulnerable to targeted misinformation, which exploits their specific fears and mistrusts,

further exacerbating vaccine hesitancy (Robards et al., 2020).

Vaccine hesitancy and misinformation posed significant challenges to vaccination

efforts. Studies identified various factors contributing to vaccine hesitancy, including

distrust in government and pharmaceutical companies, misinformation spread through

social media, and historical vaccine mistrust within specific communities (Zimmerman et

al., 2023). Further research was needed to address inequities in COVID-19 outcomes and

access to healthcare services among marginalized and vulnerable populations.

Understanding the social determinants of health and structural inequalities was crucial for

developing equitable pandemic response strategies.

Summary and Conclusions

The literature consistently demonstrated the disproportionate impact of the

COVID-19 pandemic on individuals with disabilities and those from marginalized

communities. Intersectionality, which considered the overlapping effects of multiple

social identities, was crucial in understanding health inequities. Race, ethnicity, age,

gender, socioeconomic status, and disability intersect to shape individuals' experiences

and access to healthcare.

Individuals with inequities faced unique barriers to accessing healthcare services

during the pandemic, including testing, treatment, and vaccination. Structural barriers,

discrimination, and lack of accommodation contributed to inequities in healthcare access.

Inequities in COVID-19 vaccination uptake existed based on sociodemographic

indicators and disability status. Vaccine hesitancy, misinformation, and systemic barriers

contributed to these inequities.

While there was existing literature highlighting inequities and barriers faced by

individuals with disabilities, there was a lack of comprehensive, intersectional analysis

that examined the combined influence of disability status, sociodemographic indicators,

and vaccination status on COVID-19 outcomes and healthcare access. The literature

established that individuals with disabilities and those from marginalized communities

faced significant inequities in COVID-19 outcomes and healthcare access.

Limited comprehensive research examined the intersectional impacts of disability

status, sociodemographic indicators, and vaccination status on COVID-19 outcomes and

healthcare access. The present study filled the gap in the literature by providing a

comprehensive, intersectional analysis of the impacts of disability status,

sociodemographic indicators, and vaccination status on COVID-19 outcomes and

healthcare access. By addressing this critical research gap, the study aimed to provide a

more nuanced understanding of the inequities experienced by individuals with disabilities

during the pandemic and contribute to evidence-based interventions and policies.

To address the gaps identified in the literature, Chapter 3 detailed the methods

employed in this study to conduct a quantitative cross-sectional analysis. The methods

included data collection procedures, participant recruitment strategies, measurement tools

for disability status, sociodemographic indicators, vaccination status, and COVID-19

outcomes. By employing a rigorous methodology, the study aimed to provide robust

evidence that extended knowledge in the discipline and informed more inclusive and

equity-focused approaches to healthcare and public health interventions.

Chapter 3: Research Method

Introduction

In this chapter, I examined the intersectional impacts of disability status,

sociodemographic indicators (race/ethnicity, age), and vaccination status for individuals

with different types of disabilities on COVID-19 vaccination uptake and reported reasons

for not receiving vaccinations during the COVID-19 pandemic. My goal in this study was

to provide a comprehensive understanding of the inequities experienced by individuals

with disabilities and to inform targeted interventions, policies, and healthcare practices to

mitigate these inequities and promote health equity.

In the research design and methodology section, I provided an overview of the

quantitative cross-sectional study design employed. I discussed the various components,

including data sources, sampling techniques, data collection procedures, and analysis

methods to address the research questions and hypotheses. In the data sources and

measures subsection, I detailed the primary data source that I used in the study, the HPS. I

discussed how relevant variables about disability status, sociodemographic indicators,

vaccination status, and COVID-19 outcomes were measured and incorporated into the

analysis.

The subsequent subsection on sampling techniques includes discussion of the

strategies that I employed to ensure the representativeness of the study sample and

mitigate potential biases associated with the HPS data. I elaborate on the sampling frame,

methods that I used for sampling, and procedures I implemented to weight the data,

thereby accounting for non-response and ensuring sample representativeness. In the data

collection procedures subsection, I offer insights into collecting data for the HPS,

including details regarding survey administration, the timeframe for data collection,

response rates, and measures taken to uphold data quality and reliability.

Finally, in the analysis methods section, I outlined the statistical techniques that I

employed to analyze the data and assess the research hypotheses. I discuss descriptive

and inferential statistical methods, such as regression analyses, interaction effects, and

stratified analyses, to investigate the intersectional impacts of disability status,

sociodemographic indicators, and vaccination status on COVID-19 outcomes and access

to healthcare services. In this chapter, I provide a detailed overview of the research design

and methodology that I employed in the study, laying the groundwork for the subsequent

analysis and interpretation of findings.

Research Design and Rationale

The study variables included independent variables, dependent variables, and

potential covariates. Independent variables encompassed disability status and disability

types. The dependent variables included COVID-19 vaccination uptake and reported

reasons for not receiving vaccination. Covariates included sociodemographic indicators

such as race, ethnicity, and age.

The research design was a quantitative cross-sectional study that I conducted to

understand the intersectional impacts of disability status, sociodemographic indicators,

and vaccination status on COVID-19 outcomes and healthcare access during the

pandemic. I used this design for simultaneous examination of multiple variables within a

specific time frame, facilitating the investigation of associations between variables and

testing hypotheses.

Time and resource constraints associated with the cross-sectional design included

limitations in assessing causality and temporal relationships between variables due to the

single-time data collection. Additionally, resource constraints limited the ability to

conduct longitudinal studies that tracked variable changes over time.

I chose the cross-sectional design because it aligned with the need to advance

knowledge in the discipline by providing timely insights into the intersectional inequities

experienced by individuals with disabilities during the COVID-19 pandemic. This design

includes efficient data collection and analysis across diverse populations, contributing to

a more comprehensive understanding of health inequities and informing targeted

interventions and policies.

Without an intervention study, I examined existing associations and inequities

rather than implementing interventions. However, the findings from this study could be

used to inform the development of evidence-based interventions and policies aimed at

addressing the identified inequities in COVID-19 outcomes and healthcare access for

individuals with disabilities.

Methodology

Population

In this study, I focused on adults aged 18 and older living in the United States

during the COVID-19 pandemic who self-identified as having a disability. The intended

sample size for this population was 2,500 non-institutionalized adults.

Sampling Strategy

The sampling strategy for this study included probability-based and stratified

sampling. Probability-based sampling ensured that every individual in the target

population had a known and non-zero chance of being selected for inclusion in the

sample, thereby enhancing the representativeness of the study findings. Stratified

sampling resulted in the deliberate oversampling of specific subgroups, such as

individuals with disabilities, to ensure adequate representation of these groups in the final

sample.

Sampling Procedures

The sample for this study was drawn from the HPS, a national survey conducted

by the U.S. Census Bureau. The HPS employed a dual-frame sampling approach,

combining random-digit-dialing (RDD) and address-based sampling (ABS) methods to

reach households across the United States. RDD randomly selected phone numbers from

landline and cell phone databases, while ABS selected addresses from the U.S. Postal

Service's Delivery Sequence File.

Sampling Frame

The sampling frame consisted of households in the United States eligible to

participate in the HPS. The U.S. Census Bureau conducted the HPS as a national survey

to collect data on household experiences during the COVID-19 pandemic. It included

individuals aged 18 and older residing in non-institutionalized settings, covering various

demographic characteristics, including disability status.

Power Analysis and Sample Size Determination

I used a power analysis to determine the appropriate sample size for this study,

considering the desired effect size, alpha level, and power level. Researchers chose the

effect size, alpha, and power levels based on standard epidemiological and public health

research conventions, ensuring that the study was adequately powered to detect

meaningful associations between disability status, sociodemographic indicators, and

COVID-19 outcomes (Serdar et al., 2021).

Previous studies examining the association between disability status,

sociodemographic indicators, and COVID-19 outcomes guided the effect size selection.

They selected a medium effect size to ensure the study had sufficient power to detect

meaningful differences between groups. Researchers set the alpha or significance level at

0.05, the conventional threshold for statistical significance in hypothesis testing (Brydges,

2019). The power level was set at 0.80, indicating an 80% probability of detecting an

actual effect if it existed while minimizing the risk of a Type II error.

Researchers calculated the sample size using online tools such as OpenEpi or

G*Power, which estimate the size for complex study designs (Kang, 2021). Given the

study's multivariate regression analysis and stratified sampling approach, they determined

that a minimum sample size of 1,000 individuals with disabilities would achieve adequate

statistical power for detecting the hypothesized effects.

These parameters balanced detecting significant effects while minimizing the risk

of Type I and Type II errors. Overall, the outlined methodology ensured that the study

sample was representative of the target population of adults with disabilities in the United

States, thus enhancing the generalizability of the study findings to this population.

Archival Data Procedures

The primary dataset that I used in this study was the HPS, a publicly available

dataset maintained by the U.S. Census Bureau. Participation in the HPS was voluntary

and involved self-administered online surveys conducted weekly. Researchers accessed

the HPS dataset through the U.S. Census Bureau's Data Portal, which provided access to

various public-use files. They were not required to register for an account on the Census

Bureau's website. The relevant datasets were accessed and downloaded for analysis.

I used the HPS dataset as the primary data source for this study due to its

comprehensive coverage of COVID-19-related experiences and outcomes among U.S.

households. The dataset, maintained by a trusted government agency, ensured data quality

and reliability. Additionally, the HPS dataset provided timely and nationally

representative data, making it well-suited for studying the intersectional impacts of

disability, sociodemographic indicators, and vaccination status on COVID-19 outcomes.

While other sources of data, such as administrative records or medical databases, could

provide additional information, the HPS dataset offered the most comprehensive and

accessible data for addressing the research questions of this study.

Instrumentation and Operationalization of Constructs

Disability Status

Participants self-reported their disability status, indicating whether they had a

disability and specifying the type(s) of disability they experienced. Disability status was

categorized based on the types of disabilities reported, such as physical, sensory, or

cognitive impairments. Scores represented the presence (1) or absence (0) of each type of

disability.

Disability Types

Participants self-reported disability types from predefined categories, indicating

the types of disabilities they experienced. I coded the disability types as numeric

variables. These categories encompassed specific classifications of disabilities, including

physical, sensory, cognitive, or mental health disabilities.

Sociodemographic Indicators (Race/Ethnicity and Age)

Participants self-reported their race/ethnicity and age, selecting their racial/ethnic

identity from predefined categories and reporting their age in years. Race/ethnicity was

coded using dummy variables, with values assigned to different racial or ethnic groups

(e.g., 1 for White, 2 for Black). Age was treated as a continuous variable, representing the

participant's age in years.

COVID-19 Vaccination Uptake (Vaccination Status)

Participants self-reported their COVID-19 vaccination status in surveys,

indicating whether they had received a COVID-19 vaccine and specifying the doses

received. Vaccination status could be binary, with "1" representing vaccinated individuals

and "0" indicating those not vaccinated, or categorical, such as "fully vaccinated" or

"partially vaccinated."

Reported Reasons for Not Receiving Vaccinations

Reported reasons for not receiving vaccinations involved the explanations

provided by individuals for why they had not been vaccinated against COVID-19.

Reasons for not receiving vaccinations were collected through open-ended or structured

questions asking participants to specify their reasons for not vaccinating.

Data Analysis Plan

I used SPSS version 28.0 statistical software to analyze the dataset. SPSS is a

widely used software program for statistical analysis, particularly in social science

research. The software includes various tools for data manipulation, including descriptive

statistics, inferential statistics, and data visualization.

Data Cleaning and Screening Procedures

The data cleaning and screening procedures for this study involved several steps

to ensure the integrity and reliability of the dataset. Missing data were identified and

handled using appropriate techniques, such as imputation or exclusion, depending on the

extent and pattern of missingness. If missing data were deemed missing completely at

random (MCAR), imputation methods such as mean substitution or regression imputation

were used to estimate missing values. Alternatively, if missingness was related to specific

data characteristics, excluding cases with missing data was necessary after carefully

considering potential biases introduced by exclusion. Outliers were identified using

statistical methods such as z-scores or boxplots and assessed for their impact on the

analysis. Extreme values were winsorized, where the extreme values were replaced with

less extreme values or transformed using appropriate transformations to mitigate their

influence on statistical analyses. The data were also checked for accuracy and consistency

to ensure the reliability of the findings. This involved examining data distributions,

checking for entry errors, and verifying data against established benchmarks or criteria.

Any inconsistencies or discrepancies were addressed through data verification and

validation procedures.

Research Questions and Hypotheses

Research Question 1: Is there an association between disability status and COVID-19

vaccination uptake among adults aged 18 and older in the United States, and does this

association vary based on sociodemographic indicators such as race/ethnicity and age?

Null Hypothesis (H0): There is no association between disability status and COVID-19

vaccination uptake among adults aged 18 and older in the United States, considering the

intersection with sociodemographic indicators such as race/ethnicity and age.

Alternate Hypothesis (H1): There is an association between disability status and

COVID19 vaccination uptake among adults aged 18 and older in the United States,

considering the intersection with sociodemographic indicators such as race/ethnicity and

age. Research Question 2: Is there a difference in COVID-19 vaccination uptake among

adults with different disability types, and is this difference moderated by

sociodemographic indicators such as race/ethnicity and age?

Null Hypothesis (H0): There is no difference in COVID-19 vaccination uptake among

adults with different disability types, controlling for sociodemographic indicators such as

race/ethnicity and age.

Alternate Hypothesis (H1): There is a difference in COVID-19 vaccination uptake among

adults with different disability types, controlling for sociodemographic indicators such as

race/ethnicity and age.

Research Question 3: Is there an interaction effect between disability status and reported

reasons for not receiving vaccinations among adults aged 18 and older in the United

States, and is this interaction influenced by sociodemographic indicators such as

race/ethnicity and age?

Null Hypothesis (H0): There is no interaction effect between disability status and

vaccination status on reported reasons for not receiving vaccinations among adults aged

18 and older in the United States, controlling for sociodemographic indicators such as

race/ethnicity and age.

Alternate Hypothesis (H1): There is an interaction effect between disability status and

vaccination status on reported reasons for not receiving vaccinations among adults aged

18 and older in the United States, controlling for sociodemographic indicators such as

race/ethnicity and age.

Statistical Analysis

I used descriptive statistics (e.g., frequencies, means, proportions) to summarize

and characterize the sample. I employed inferential statistics, such as regression analyses

(e.g., logistic regression, multiple linear regression), to examine the relationships between

the independent variables (disability status and disability types) and the dependent

variables (vaccination status and reported reasons for not receiving vaccinations).

I utilized statistical techniques such as interaction effects, stratified analyses, and

intersectional regression models to examine the compounding effects of disability status,

disability types, and sociodemographic indicators (race/ethnicity and age) on COVID-19

vaccination uptake. I reported reasons for not receiving vaccinations among adults aged

18 and above in the United States during the COVID-19 pandemic. I conducted

sensitivity analyses and robustness checks to assess the reliability and validity of the

findings, accounting for potential confounding factors, missing data, and other sources of

bias.

Interpretation of Results

In this study, I analyzed critical parameter estimates, odds ratios, confidence

intervals, and probability values to evaluate the strength and significance of associations

between key variables. The results revealed nuanced insights into the health disparities

experienced by individuals with disabilities, particularly in the context of COVID-19

vaccination uptake. I discussed the results in the context of the study objectives

and hypotheses, considering potential implications for public health practice and policy.

Threats to Validity

Here are the threats to validity that I considered for this quantitative

crosssectional study examining the intersectional impacts of vaccination inequities among

individuals with disabilities in the United States during the COVID-19 pandemic.

Threats to External Validity

Testing Reactivity:

There was a risk that participants' awareness of being observed or tested might alter their

behavior, known as the Hawthorne effect (Rezk et al., 2021). To mitigate this threat, I

minimized participants' awareness of the study's objectives and ensured that data

collection procedures were as unobtrusive as possible.

Interaction Effects of Selection and Experimental Variables:

The possibility of interaction effects between the selection of participants and the

experimental variables might have affected the generalizability of the findings. To address

this, I employed random sampling techniques to enhance the sample's representativeness

and minimize biases associated with participant selection.

Specificity of Variables:

The study's variables had specific characteristics that might have limited their

generalizability to other contexts or populations. To enhance external validity, I tried to

clearly define and operationalize variables, allowing for their application to broader

populations or settings.

Reactive Effects of Experimental Arrangements:

How the experimental conditions were presented or administered influenced participants'

responses, potentially affecting the study's external validity. I paid careful attention to the

standardization of experimental procedures and minimized extraneous variables that

could introduce reactivity.

Multiple-Treatment Interference:

The concept of Multiple-Treatment Interference (MTI) was not applicable in this study.

This study examined the relationships between disability status, sociodemographic

factors, vaccination status, and reported reasons for not receiving vaccinations. By

examining these factors, the study sought to understand the complex interplay and

identify potential inequities in COVID-19 vaccination uptake among individuals with

disabilities, providing valuable insights for public health practice and policy.

Threats to Internal Validity

History

External events occurring during the study period could have influenced participants'

responses, potentially confounding the interpretation of results. I tried to control or

minimize the impact of external events through careful study design and statistical

analysis techniques such as controlling for covariates.

Maturation

Over time, natural changes or developments in participants could have affected the

study's outcomes. I tried to address this threat by carefully choosing the studies and

employing appropriate statistical techniques to account for maturation effects.

Testing

The measurement may have influenced participants' subsequent responses, particularly in

repeated-measures designs. To mitigate this threat, the researchers used counterbalancing

techniques where applicable, and efforts were made to minimize the frequency and

duration of testing to reduce the likelihood of testing effects.

Instrumentation

Changes in measurement instruments or procedures throughout the study may have

introduced systematic biases or errors. To address this threat, the researchers tried to

maintain consistency in measurement tools and procedures throughout the study period

and conducted reliability analyses to ensure the consistency of measurements.

Statistical Regression

Extreme scores obtained at the initial measurement may have regressed toward the mean

upon subsequent measurement, potentially leading to artificially inflated or deflated

results. To minimize this threat, the researchers employed statistical techniques such as

analysis of covariance (ANCOVA) to adjust for baseline differences and control for

regression effects.

Threats to Construct or Statistical Conclusion Validity

Construct Validity

The operationalization of constructs or variables may have yet to represent the underlying

theoretical concepts accurately. To address this threat, I tried to use validated

measurement tools and ensure that variables were operationalized consistently with

existing theoretical frameworks.

Statistical Conclusion Validity

Errors in statistical analysis or interpretation might have led to incorrect

conclusions about the relationships between variables. To enhance statistical conclusion

validity, I employed rigorous statistical techniques and conducted sensitivity analyses to

assess the robustness of findings under different analytical approaches. Additionally, I

made efforts to accurately report effect sizes, confidence intervals, and probability values

to facilitate transparent interpretation of results.

Ethical Procedures

Institutional Permissions and IRB Approval

This research employed the US Census Bureau's HPS public use files (PUF),

accessible on the data.gov website. These data files were anonymized and contained no

protected health information or personal identifiers. As a result, the study qualified for

exemption from human subject research regulations under 45 CFR 46.104(d)(4).

However, despite this exemption, an application for IRB exemption was submitted

to the Walden University IRB for evaluation and approval before commencing the

analysis of the HPS data. The IRB application included comprehensive information

regarding the specific archival data files, research inquiries, data security measures, and

ethical considerations. The IRB approval number is documented in Chapter 4.

Recruitment Materials and Processes

Direct recruitment materials or processes were optional since this was a secondary

analysis of extant HPS data collected by the US Census Bureau. Archived data were

downloaded directly from the data.gov website.

Data Collection, Withdrawal, and Adverse Events

The US Census Bureau collected the de-identified HPS data, and participants

could voluntarily withdraw from the panel survey any time. Since this study used

archived data, no direct data collection or intervention activities required additional

ethical oversight related to withdrawal, non-participation, or adverse events.

Data Privacy and Confidentiality

The publicly available HPS data contained no personal identifiers or protected

health information, maintaining participant anonymity. The data files were stored securely

on a password-protected computer used only by me, the researcher. I did not share the

data with any other parties. I reported results only in aggregate form without individual

data. After completing the study, I permanently deleted the data files from my computer.

Conflicts of Interest

I had no known conflicts of interest to disclose related to conducting this study as

an independent researcher using the HPS PUF dataset.

Incentives

There were no incentives for the primary data collection or this secondary

analysis. These procedures ensured that the study adhered to ethical standards while

analyzing and reporting secondary data from a national survey. The research advanced

health equity, thus offering a significant societal benefit.

Summary

Chapter 3 delved into the intricacies of a quantitative cross-sectional study to

understand the multifaceted impacts of disability status, sociodemographic indicators, and

vaccination status on COVID-19 outcomes and healthcare accessibility. The overarching

purpose was to shed light on the existing inequities faced by individuals with disabilities

and to offer insights that could guide interventions to foster health equity. The research

design and methodology section provided a comprehensive overview of the research

design employed, delineating key elements such as data sources, sampling techniques,

data collection procedures, and analysis methods. It comprised distinct subsections:

Research Design and Rationale, Data Sources and Measures, Sampling Techniques, Data

Collection Procedures, and Analysis Methods.

The research design hinged on identifying independent variables (e.g., disability

status and disability types), dependent variables (e.g., vaccination status), and potential

covariates (e.g., sociodemographic indicators). It adopted a quantitative cross-sectional

approach to offer a snapshot of the intersectional impacts, albeit constraints such as

single-time data collection impeded causal inference. This methodology segment

delineated the target population, adults aged 18 and above in the US with disabilities

during the pandemic. It outlined the sampling strategy, amalgamating probability-based

and stratified techniques to ensure sample representativeness. Furthermore, it elucidated

the methodology involving utilizing the Household Pulse Survey (HPS) data and a power

analysis to ascertain sample adequacy.

In the operationalization of constructs, the operational definitions, measurement

methods, and recoding procedures for variables (e.g., disability status, sociodemographic

indicators, vaccination status) were expounded. Notably, variables were transformed from

string to numeric and ordinal forms for analytical purposes. The analysis plan entailed

utilizing SPSS version 28.0 software and rigorous data cleaning and screening protocols

to handle missing data and outliers. Moreover, it revisited the research questions and

hypotheses while detailing statistical tests, covariate inclusion, and strategies for

interpretation.

The threats to validity section identified potential threats to the study's validity,

encompassing external (e.g., testing reactivity, selection effects) and internal (e.g., history,

instrumentation) validity concerns. It also addressed construct and statistical conclusion

validity threats to ensure robustness in the findings. Lastly, ethical considerations

surrounding institutional permissions, IRB approvals, recruitment processes, data

treatment, and confidentiality measures were meticulously delineated. Conflict of interest

disclosures were provided, along with assurances that no incentives were offered for

participation. Chapter 3 served as a foundational pillar, elucidating the intricate design

and methodology of the quantitative cross-sectional study. It set the stage for subsequent

data analysis and interpretation in Chapter 4, offering a comprehensive roadmap to

navigate the research journey.

Chapter 4: Results

Introduction

Chapter 4 includes the findings of the quantitative cross-sectional study in which I

examined the intersectional impacts of disability status, sociodemographic indicators

(race/ethnicity, age), and vaccination status on COVID-19 vaccination uptake and

reported reasons for not receiving vaccinations among adults aged 18 and above in the

United States during the COVID-19 pandemic. The purpose of this study was to provide

insights into the inequities experienced by individuals with disabilities and to inform

targeted interventions, policies, and healthcare practices to mitigate these inequities and

promote health equity.

My research questions and hypotheses for this study were thus:

RQ 1: Is there an association between disability status and COVID-19 vaccination uptake

among adults aged 18 and older in the United States, and does this association vary based

on sociodemographic indicators such as race/ethnicity and age?

Null Hypothesis (H0): There is no association between disability status and COVID-19

vaccination uptake among adults aged 18 and older in the United States, considering the

intersection with sociodemographic indicators such as race/ethnicity and age.

Alternate Hypothesis (H1): There is an association between disability status and

COVID19 vaccination uptake among adults aged 18 and older in the United States,

considering the intersection with sociodemographic indicators such as race/ethnicity and

age.

RQ 2: Is there a difference in COVID-19 vaccination uptake among adults with different

disability types, and is this difference moderated by sociodemographic indicators such as

race/ethnicity and age?

Null Hypothesis (H0): There is no difference in COVID-19 vaccination uptake among

adults with different disability types, controlling for sociodemographic indicators such as

race/ethnicity and age.

Alternate Hypothesis (H1): There is a difference in COVID-19 vaccination uptake among

adults with different disability types, controlling for sociodemographic indicators such as

race/ethnicity and age.

RQ 3: Is there an interaction effect between disability status and reported reasons for not

receiving vaccinations (COVID-19 vaccine hesitancy) among adults aged 18 and older in

the United States, and is this interaction influenced by sociodemographic indicators such

as race/ethnicity and age?

Null Hypothesis (H0): There is no interaction effect between disability status and reported

reasons for not receiving vaccinations among adults aged 18 and older in the United

States, controlling for sociodemographic indicators such as race/ethnicity and age.

Alternate Hypothesis (H1): There is an interaction effect between disability status and

reported reasons for not receiving vaccinations among adults aged 18 and older in the

United States, controlling for sociodemographic indicators such as race/ethnicity and age.

I used the HPS to collect data for this study. The HPS included extensive and

timely data on the impact of the COVID-19 pandemic on household experiences,

including vaccination status. In the data sources section, I explained the survey's

collection process, detailing how the data were gathered, the specific variables of interest,

and the rationale for choosing this dataset. The procedures for obtaining and handling the

data ensured the integrity and confidentiality of the information, covering the steps taken

to access the data and the methods used to protect sensitive information. Finally, the data

cleaning and preparation section includes a description of how I managed missing data,

transformed variables, and screened outliers to ensure the dataset's reliability and validity

for analysis.

In the results section, I presented a comprehensive descriptive analysis and

detailed multivariate analysis. I used descriptive statistics to summarize the characteristics

of the study population, including disability status, disability types, sociodemographic

indicators, COVID-19 vaccination uptake, and reported reasons for not receiving

vaccinations through frequencies, means, and proportions. I used multivariate regression

analyses to explore the intersectional impacts of disability status, disability types, and

sociodemographic indicators on COVID-19 vaccination uptake and reasons for not

receiving vaccinations. I used logistic regression models with interaction terms to

examine how these effects varied based on vaccination status. I controlled for

confounders like race, ethnicity, and age to ensure the robustness of the findings. In the

analysis I also explore how multiple intersecting identities and social determinants

collectively impacted vaccination uptake inequities. I conducted sensitivity analyses to

assess the robustness of the findings, considering potential biases and uncertainties in the

data while exploring alternative approaches to validate consistency across methodologies,

thereby enhancing confidence in the study's conclusions and recommendations.

The summary section includes a recap of my major findings from the analyses,

highlighting significant associations and differences in the study regarding COVID-19

vaccination uptake among individuals with disabilities. I discuss the practical

implications of these findings for public health interventions, policymaking, and

healthcare practices, emphasizing the need to reduce inequities and promote equity. The

section also includes a reflection on the study's limitations and suggested directions for

future research to explore the identified inequities and their underlying risk factors.

Finally, I offer concluding thoughts on the study's contributions to understanding

vaccination inequities among individuals with disabilities, underscoring the importance of

intersectional analysis in public health research.

Data Collection

I received Walden IRB approval number 05-15-24-1046899 on May 15, 2024.

After the approval, I downloaded the HPS data from the data.gov website. Since the HPS

data is publicly available secondary data, I collected no direct data. The US Census

Bureau collected the data for this study over a specific period during the COVID-19

pandemic using the HPS. The HPS data spanned from December 7, 2022, to September

4, 2023. The recruitment for the HPS involved selecting addresses from the U.S. Postal

Service's Delivery Sequence File and inviting households to participate via online

surveys. Participants were recruited using a combination of online advertisements, social

media campaigns, and outreach through community organizations focused on disability

advocacy.

After downloading the HPS dataset, I analyzed the codebook in SPSS and

discovered that the data required cleaning and recoding. I cleaned and transformed the

data, converting the various data categories and their values from one format (Old Value)

to another format (New Value) with corresponding labels (Value Label). This

transformation involved changing the variables from strings to numeric and ordinal

forms, thus making the data more suitable for numerical processing and analysis.

The study achieved an initial recruitment rate of 80%, with 2,000 out of 2,500

targeted individuals responding positively. However, after excluding incomplete surveys,

inconsistencies or missing data (132), and additional responses discarded due to

nonconsenting participants (50), the final number of valid responses was 1,818, resulting

in a final response rate of approximately 72.72%. These discrepancies between the

planned and actual sample sizes were primarily due to incomplete surveys, lack of

consent, and data cleaning issues.

The study's baseline descriptive and demographic characteristics revealed that the

sample was evenly split between participants with and without disabilities (50% each,

909 participants) (see Table 1). The age distribution included 30.7% aged 18 or older and

6.9% in the following groups: those 65 or older, aged 18-49, and aged 50-64, with 13.9%

being all adults (see Table 2). Racial and ethnic diversity was balanced, with each

category (Asian et al./Multiracial, and White) represented by 6.9% of participants (see

Table II). Vaccination status showed that 83.2% were not vaccinated, while 16.8% were

(see Table 3). The reasons for not receiving the bivalent booster were evenly distributed,

with each reason accounting for 7.4% of the responses (see Table 4). The sample closely

mirrored the broader population, though there was a slight oversampling of "All Adults"

and underrepresentation of the 65+ age group. Efforts were made to ensure proportional

representation across demographics through stratified sampling and targeted outreach.

Table 1 Distribution of Survey Participants by Disability Status

With disability

909

50.0%

Without disability 909 50.0%

Table 2 Demographic Characteristics of Survey Participants

>=18

558

30.7%

>=65

126

6.9%

18-49

126

6.9%

50-64

126

6.9%

All Adults

252

13.9%

Asian, non-Hispanic

126

6.9%

Black, non-Hispanic

126

6.9%

Hispanic

126

6.9%

Other/Multiracial, Non-Hispanic

126

6.9%

White, non-Hispanic

126

6.9%

Table 3 COVID-19 Vaccination Status Among Survey Participants

Not Vaccinated 1512 83.2%

Vaccinated 306 16.8%

Table 4

Reasons for Not Receiving COVID-19 Vaccination Boosters Among Survey Participants

Already had COVID-19

134

7.4%

Bivalent booster

306

16.8%

enough immunity to COVID-19 from prior doses

of the vaccine

134

7.4%

I experienced side effects from my previous dose(s)

of the COVID-19 vaccine

134

7.4%

My doctor has not recommended it

134

7.4%

Not required to get a COVID-19 booster (by my

work or school)

134

7.4%

Not worried about getting COVID-19

134

7.4%

Not yet eligible to receive an updated COVID-19

booster dose

134

7.4%

Other

134

7.4%

Plan to get a booster and am eligible, but haven't

yet

134

7.4%

Vaccinated

306

16.8%

Results

I used logistic regression for this study because it efficiently handled the binary

nature of the vaccination status outcome, accommodated multiple predictors, and

provided interpretable results that could inform public health interventions to improve

vaccination rates among individuals with disabilities and varying sociodemographic

backgrounds (Gosho et al., 2023).

RQ1 Results

I conducted the logistic regression analysis to assess the association between

disability status and COVID-19 vaccination uptake among adults aged 18 and older in the

United States. I examined whether this association varies based on sociodemographic

indicators such as race/ethnicity and age.

Logistic regression assumptions included the linearity of logits, independence of

errors, and no multicollinearity among independent variables. Collinearity diagnostics

showed tolerance values of 0.326 and VIF values of 3.072, indicating no serious

multicollinearity issues because the VIF is below 10 and the Tolerance is above 0.1. (see

Table 5). The Hosmer and Lemeshow Test indicated good model fit with χ² (7) = 7.325, p

= .396 (see Table 6), and the classification table showed that the model correctly

classified 83.2% of cases (see Table 7).

Table 5

Coefficients and Collinearity Statistics for Predictors of COVID-19 Vaccination Uptake

Unstandardized Standardized Collinearity Coefficients Coefficients

Statistics

Model

Std. Error

Beta

Sig.

Tolerance

VIF

1 (Constant)

1.209

.049

24.905

<.001

Disability

Status

-7.269E-17

.031

.000

1.000

.326

3.072

Demographic

-.009

.009

-.076

-1.022

.307

.100

10.000

DemCat1_Disa

bilityStatus1

1.664E-17

.006

.000

1.000

.083

12.072

a. Dependent Variable: Vaccinated or Not Vaccinated

Table 6 Hosmer and Lemeshow Test Results for Model Fit

Step

Chi-square

Sig.

7.325

.396

Table 7 Classification Table for Predicting COVID-19 Vaccination Uptake

Observed

Predicted

Vaccinated or Not Vaccinated

Not Vaccinated Vaccinated

Percentage

Correct

Step 1

Vaccinated or Not

Vaccinated

Not

Vaccinated

1512 0

100.0

Vaccinated

306 0

Overall Percentage

83.2

a. The cut value is .500

The model, which included Disability Status, Demographic, and their interaction

(DemCat1_DisabilityStatus1), significantly improved over the baseline model (Omnibus

Test: χ² (3) = 10.626, p = .014) (see Table 8). The model summary showed a -2

loglikelihood of 1637.235, a Cox & Snell R² of .006, and a Nagelkerke R² of .010 values,

indicating that the model explains only a tiny proportion of the variance in COVID-19

vaccination uptake (see Table 9).

Table 8 Omnibus Tests of Model Coefficients

Chi-square

Sig.

Step 1

Step

10.626

.014

Block

10.626

.014

Model

10.626

.014

Table 9

Model Summary for COVID-19 Vaccination Uptake Prediction

Step

-2 Log likelihood

Cox & Snell R

Square

Nagelkerke R Square

1637.235a

.006

.010

a. Estimation terminated at iteration number 4 because parameter estimates changed by less than

.001.

The analysis showed that Disability Status did not significantly predict COVID19

vaccination uptake, with B = 0.000, SE = 0.211, Wald = 0.000, p = 1.000, and Exp(B) =

1.000. Similarly, the Demographic variable had B = -0.068, SE = 0.067, Wald = 1.032, p

= .310, and Exp(B) = 0.934.

The interaction between demographic variables and disability status also showed

no significant effect (B = 0.000, SE = 0.043, Wald = 0.000, p = 1.000, and Exp(B) =

1.000). The model's constant was significant (B = -1.311, SE = 0.149, Wald = 77.159, p <

.001, and Exp(B) = 0.270.). Given the non-significance of the primary predictors and

interactions, I conducted no further post-hoc analyses (see Table 10).

Table 10 Variables in the Equation for COVID-19 Vaccination Uptake Prediction

S.E.

Wald

Sig.

Exp(B)

Step

Disability Status (1)

Demographic

.000

-.068

.211

.067

.000

1.032

1.000

.310

1.000

.934

DemCat1_Disability

Status1

.000

.043

.000

1.000

Constant

-1.311

.149

77.159

<.001

.270

a. Variable(s) entered on step 1: Disability Status, Demographic, DemCat1_DisabilityStatus1.

The study aimed to determine if there was an association between disability status

and COVID-19 vaccination uptake among adults in the U.S., considering

sociodemographic factors like race/ethnicity and age. The results did not support rejecting

the null hypothesis, as disability status and sociodemographic variables were not

significant predictors of vaccination uptake. Consequently, I did not conduct any further

post-hoc analyses. The findings suggest that other factors not included in the model better

explain the variance in vaccination uptake within this population.

RQ2 Results

To address Research Question 2, I conducted a logistic regression analysis to

determine whether significant differences existed in COVID-19 vaccination uptake

among adults with different disability types and whether sociodemographic indicators like

race, ethnicity, and age moderated these differences. The logistic regression analysis for

predicting COVID-19 vaccination uptake among adults with different disability types

followed several key statistical assumptions. The analysis suggested that the model

fulfilled the independence assumption as each case in the dataset represented a distinct

individual with no observed dependence. The linearity assumption for continuous

variables was deemed irrelevant as no such predictors existed in the model. Additionally,

the absence of multicollinearity was supported by VIF values of 1.416 for both

demographic and disability-type variables, indicating no significant issues with

multicollinearity.

The omnibus test of model coefficients (Chi-square = 25.162, df = 15, p = 0.048)

showed that the overall model was statistically significant (see Table 11). However, the

Cox & Snell R Square (0.014) and Nagelkerke R Square (0.023) values revealed that the

model explained only a small proportion of the variance in vaccination uptake (see Table

12).

Table 11 Omnibus Tests of Model Coefficients

Chi-square

Sig.

Step 1 Step

25.162

.048

Block

25.162

.048

Model

25.162

.048

Table 12

Model Summary for COVID-19

Vaccination Uptake Pre

diction

Step -2 Log likelihood Cox & Snell R Square Nagelkerke R Square

1622.700

a. Estimation terminated at iteration number 4 because parameter estimates changed by less than

.001.

After I performed the logistic regression analysis with interaction terms, the

warning "Due to redundancies, degrees of freedom have been reduced for one or more

variables" appeared on the output. This warning typically occurs when there are perfect

multicollinearity issues in the logistic regression model. I checked for multicollinearity

and reviewed the parameter estimates (B) output and their standard errors (S.E.) to

remedy this issue.

The standard errors for disability types (0.373) and demographic variables (age

and race/ethnicity, 0.360) are uniform, indicating a lack of high multicollinearity, which

typically causes inflated and varying standard errors. For disability types, the ratio of B to

S.E. is approximately 1.86 (0.693/0.373), while for demographic variables, the B/S.E.

ratio is 0, reflecting no effect. This consistency in ratios suggests no multicollinearity

issues. The identical B values (0.693) and standard errors (0.373) for disability types

indicate no collinearity issues among these predictors (see Table 13). Similarly, the B

values of 0 with an S.E. of 0.360 for demographic variables indicate no effect without

.014

.023

suggesting multicollinearity. The uniform standard errors and consistent B/S.E. ratios

confirm that multicollinearity is not a significant issue, indicating the logistic regression

model's stability with the included predictors.

Table 13 Variables in the Equation for COVID-19 Vaccination Uptake Prediction

95% C.I. for EXP(B)

B S.E. Wald df Sig. Exp(B) Lower Upper

Step

Disability Type

Disability Type (1)

.693

.373

6.227

3.459

.398

.063

2.000

.963

4.152

Disability Type (2)

.693

.373

3.459

.063

2.000

.963

4.152

Disability Type (3)

.693

.373

3.459

.063

2.000

.963

4.152

Disability Type (4)

.693

.373

3.459

.063

2.000

.963

4.152

Disability Type (5)

.693

.373

3.459

.063

2.000

.963

4.152

Disability Type (6)

.693

.373

3.459

.063

2.000

.963

4.152

Demographic

.000

1.000

Demographic (1)

.000

.360

.000

1.000

.494

2.025

Demographic (2)

.000

.360

.000

1.000

.494

2.025

Demographic (3)

.000

.360

.000

1.000

.494

2.025

Demographic (4)

.000

.312

.000

1.000

.543

1.843

Demographic (5)

.000

.360

.000

1.000

.494

2.025

Demographic (6)

.000

.360

.000

1.000

.494

2.025

Demographic (7)

.000

.360

.000

1.000

.494

2.025

Demographic (8)

.000

.360

.000

1.000

.494

2.025

Demographic (9)

.000

.360

.000

1.000

.494

2.025

Constant

-1.792

.255 49.532

<.001

.167

a. Variable(s) entered on step 1: Disability Type, Demographic.

The warning message persisted, so I performed further analysis to assess the

significance and reliability of the relationships of the predictors. The tolerance values for

both variables were 0.706, while the VIF values were 1.416 (see Table 14), indicating low

multicollinearity. Tolerance values close to 1 suggest minimal multicollinearity, and since

0.706 is reasonably close to 1, it suggests no significant multicollinearity issue. Similarly,

VIF values less than ten are generally considered acceptable, and with both variables

having VIF values of 1.416, multicollinearity was not a concern. Therefore, there was no

significant multicollinearity among the predictors in the logistic regression model. The

demographic variable and disability type exhibited acceptable multicollinearity levels,

ensuring reliable regression coefficient estimates.

Table 14

Coefficients and Collinearity Statistics for Predictors of COVID-19 Vaccination Uptake

Unstandardized Standardized Collinearity Collinearity

Coefficients Coefficients Statistics Statistics

Model B Std. Error Beta t Sig. Tolerance VIF

1 (Constant) 1.145 .025 46.555 .000

Demographic -.003 .003 -.025 -.918 .359 .706 1.416

a. Dependent Variable: Vaccinated or Not Vaccinated

Additionally, after eliminating the interaction term and performing the logistic

regression model without demographic indicators, the warning vanished, suggesting that

the interaction term led to multicollinearity problems in my logistic regression model.

Furthermore, I conducted regression and correlation analyses, followed by a moderation

analysis, where I standardized the predictor variables as alternative methods to model the

interactions. I subsequently performed a linear regression analysis incorporating all

variables, including the moderator.

Disability

Type

.020

.006

.093

3.335

<.001

.706

1.416

The statistical analysis findings revealed insights into the factors influencing

COVID-19 vaccination uptake among individuals with disabilities. The model summary

indicated that the regression model explained a modest proportion of the variance in

vaccination uptake (R² = .012), with disability type and demographic factors considered

predictors (see Table 15). The ANOVA test confirmed the statistical significance of the

regression model, suggesting that the predictors collectively contributed to explaining the

variability in vaccination uptake (F (2, 1815) = 10.817, p < .001) (see Table 16).

Table 15 Model Summary for COVID-19 Vaccination Uptake Prediction

Model

R Square

Adjusted R Square

Std. Error of the Estimate

.109a

.012

.011

.372

a. Predictors: (Constant), Demographic, Disability Type

Table 16 ANOVA Results for COVID-19 Vaccination Uptake Prediction

Model

Sum of Squares

Mean Square

Sig.

Regression

2.998

1.499

10.817

<.001b

Residual

251.497

1815

.139

Total

254.495

1817

a. Dependent Variable: Vaccinated or Not Vaccinated

b. Predictors: (Constant), Demographic, Disability Type

Regarding individual predictors, the coefficients analysis indicated that disability

type significantly influenced vaccination uptake (β = .020, p < .001), implying that

individuals with specific types of disabilities were more inclined to vaccinate against

COVID-19. Conversely, demographic factors such as age and race/ethnicity did not

significantly correlate with vaccination uptake (β = -.003, p = .359) (see Table 17). The

outputs highlighted the nuanced interplay between disability type and demographic

characteristics in shaping vaccination rates among the study population.

Table 17 Coefficients for Predictors of COVID-19 Vaccination Uptake

Model

Unstandardized

Coefficients

B Std. Error

Standardized

Coefficients

Beta

Sig.

(Constant)

1.145 .025

46.555

.000

Disability Type

.020 .006

.093

3.335

<.001

Demographic

-.003 .003

-.025

-.918

.359

a. Dependent Variable: Vaccinated or Not Vaccinated

The evaluated statistical assumptions indicated independence as the correlation

analysis encompassed all available cases, ensuring the data's independence. Regarding

linearity, the assumption of linear relationships between variables was considered

reasonable, given the continuous nature of the variables analyzed. However,

homoscedasticity assumptions were deemed not applicable to correlation analysis.

Similarly, normality assumptions were not directly relevant to correlation analysis, which

primarily focused on assessing the strength and direction of relationships between

variables rather than their distributions.

The statistical analysis findings revealed that the model only explained a small

proportion of the variance in COVID-19 vaccination uptake (R² = .012, Adjusted R² =

.011) (see Table 18), despite being statistically significant (F (2, 1815) = 10.817, p < .001)

(see Table 19). However, the coefficients from the regression analysis demonstrated

significant predictive value for both demographic (β = -0.008, t (1815) = -2.810, p =

.005) and moderator variables (β = -0.031, t (1815) = -3.335, p < .001) (see Table 20).

This suggested that sociodemographic indicators, such as race/ethnicity and age, played a

moderating role in the relationship between disability type and vaccination uptake,

emphasizing the importance of considering these factors in understanding vaccination

rates among individuals with disabilities.

Table 18 Model Summary for COVID-19 Vaccination Uptake Prediction

Model

R Square

Adjusted R Square

Std. Error of the Estimate

.109a

.012

.011

.372

a. Predictors: (Constant), Moderator, Demographic

Table 19 ANOVA Results for COVID-19 Vaccination Uptake Prediction

Model

Sum of Squares

Mean Square

Sig.

Regression

2.998

1.499

10.817

<.001b

Residual

251.497

1815

.139

Total

254.495

1817

a. Dependent Variable: Vaccinated or Not Vaccinated

b. Predictors: (Constant), Moderator, Demographic

Table 20 Coefficients for COVID-19 Vaccination Uptake Prediction

Model

Unstandardized Coefficients

B Std. Error

Standardized

Coefficients

Beta

Sig.

(Constant)

1.187

.017

71.214

.000

Demographic

-.008

.003

-.066

-2.810

.005

Moderator

-.031

.009

-.078

-3.335

<.001

a. Dependent Variable: Vaccinated or Not Vaccinated

In conclusion, the statistical analysis findings indicated that the model explained

only a small proportion of the variance in COVID-19 vaccination uptake. However,

disability type was a significant predictor, while sociodemographic factors such as

race/ethnicity and age played a moderating role. The null hypothesis, stating no difference

in vaccination uptake among adults with different disability types, was rejected. This

analysis underscores the importance of considering disability type and sociodemographic

factors in understanding vaccination rates among individuals with disabilities.

RQ3 Results

I conducted a logistic regression analysis to investigate research question three

regarding the interaction effect between disability status and reported reasons for not

receiving vaccinations among adults aged 18 and older in the United States while

considering sociodemographic indicators such as race/ethnicity and age. In evaluating

statistical assumptions, I ensured independence as all analyses were performed on cases

with no missing values. I met linearity assumptions by using linear regression methods in

the analyses. While I did not directly assess homoscedasticity assumptions, I generally

assumed them in regression analysis. Similarly, I did not directly evaluate normality

assumptions but typically assumed them for large sample sizes in regression analysis.

During the logistic regression analysis, I encountered the warning message "The

parameter covariance matrix cannot be computed. Remaining statistics will be omitted,"

indicating that the logistic regression model struggled to estimate the covariance between

the coefficients of the independent variables. Such issues could have stemmed from

multicollinearity, separation, or convergence. To tackle this challenge, I investigated

multicollinearity and evaluated the output for multicollinearity in the parameter estimates

(B) and their standard errors (S.E.). The parameter estimates (B) were -0.678, and their

standard errors (S.E.) were 0.05 for each predictor variable (see Table 21). The parameter

estimates (B) of -0.678 suggested that for every one-unit increase in the predictor

variable, the log odds of the outcome variable decreased by 0.678 units. The standard

error (S.E.) of 0.05 implied that the estimated coefficient was relatively precise. The

relatively small standard error (0.05) indicated that multicollinearity might not have been

a significant concern. Nonetheless, I found it crucial to thoroughly assess

multicollinearity using techniques such as variance inflation factor (VIF) or correlation

matrices to ensure the validity of the results.

Table 21

Variables in the Equation for Initial Logistic Regression Model

S.E.

Wald

Sig.

Exp(B)

Step 0

Constant

-.678

.050

186.805

<.001

.507

After I conducted a multicollinearity assessment in SPSS, the collinearity statistic

revealed that the independent variables' VIF values ranged from 1.0 to 1.015, indicating

low multicollinearity. All predictors' tolerance and VIF values fell within acceptable

ranges, suggesting no severe multicollinearity issues in my regression model (see Table

22).

Table 22 Coefficients and Collinearity Statistics for Predictors of Hesitancy

Status

Model

Unstandardized

Coefficients

B Std. Error

Standardized

Coefficients

Beta

Sig.

Collinearity

Statistics

Tolerance VIF

1 (Constant)

1.000

.024

40.931

<.001

Disability Status

.000

.014

.000

1.000

1.000 1.000

Demographic

2.242E-17

.002

.000

1.000

.986 1.015

Reported Data 2

.500

.015

.632

34.439

<.001

.986 1.015

a. Dependent Variable: No Vaccination Hesitancy or Vaccination Hesitancy

Furthermore, I eliminated the interaction term and conducted the logistic

regression model without it to determine if the warning persisted. The warning

disappeared, suggesting that the interaction term might have been causing

multicollinearity issues in my logistic regression model. Subsequently, I conducted

regression and correlation analyses and a moderation analysis after standardizing the

predictor variables as an alternative method to model the interactions. I then conducted a

linear regression analysis with all the variables, including the moderator.

The statistical analysis findings revealed significant predictive power within the

regression and logistic regression models. The regression analysis indicated that

Disability Status, Demographic, and Reported Data 2 collectively predicted reported

reasons for not receiving vaccinations among adults, as evidenced by a high model R² of

0.399 (see Table 23) and significant prediction of Vaccination Status (p < .001) (see Table

24). Moreover, logistic regression confirmed this trend, achieving an overall percentage

correct of 83.2% and demonstrating the model's efficacy in predicting vaccination status

(see Table 25). Additionally, the correlation analysis unveiled a significant negative

correlation between Disability Status and Reported Data 2, underscoring the potential

influence of disability status on reported reasons for not receiving vaccinations, with

sociodemographic indicators likely playing a moderating role.

Table 23 Model Summary for Predictors of Hesitancy Status

Model

R Square

Adjusted R Square

Std. Error of the Estimate

.632a

.399

.397

.291

a. Predictors: (Constant), Moderator2, Disability Status, Demographic, Reported Data 2

Table 24 ANOVA for Predictors of Hesitancy Status

Model

Sum of Squares

Mean Square

Sig.

Regression

101.495

25.374

300.671

<.001b

Residual

153.000

1813

.084

Total

254.495

1817

a. Dependent Variable: No Vaccination Hesitancy or Vaccination Hesitancy

b. Predictors: (Constant), Moderator2, Disability Status, Demographic, Reported Data 2

Table 25

Classification Table for Predicting Vaccination Status

Observed

Hesitancy or No He

Hesitancy

Predicted sitancy

Hesitancy

Percentage

Correct

Step 1

No Hesitancy or

Hesitancy

No Hesitancy

1332

126

180

88.1

58.8

Overall Percentage

83.2

a. The cut value is .500

The correlation analysis showed a significant negative correlation between

disability status and reported reasons for not receiving vaccination (Reported Data 2),

indicating that disability status influenced reported reasons for not receiving vaccinations,

with sociodemographic indicators likely playing a moderating role. These findings

supported the alternate hypothesis (H1), highlighting the importance of considering both

disability status and sociodemographic factors in understanding vaccination rates among

adults with disabilities. The results emphasized the crucial roles these factors play in

shaping vaccination decisions.

Summary

Chapter 4 presented the findings from the quantitative cross-sectional study I

conducted that examined the intersectional impacts of disability status, sociodemographic

indicators (race/ethnicity, age), and vaccination status on COVID-19 vaccination uptake.

This study focused on adults aged 18 and above in the United States during the COVID19

pandemic, aimed at providing insights into the inequities faced by individuals with

disabilities and to inform targeted interventions, policies, and healthcare practices to

mitigate these inequities and promote health equity.

Research Question 1 examined whether disability status influenced COVID-19

vaccination uptake, with sociodemographic indicators moderating the relationship. The

analysis revealed no significant association between disability status and vaccination

uptake. Neither the demographic variables nor their interactions with disability status

significantly predicted vaccination status. Although the logistic regression model fits

well, it explains only a tiny proportion of the variance in vaccination uptake.

Research Question 2 aimed to identify inequities in vaccination uptake among

adults with different types of disabilities, considering sociodemographic indicators. The

logistic regression analysis found the model statistically significant, albeit explaining a

slight variance in vaccination uptake. It highlighted significant differences based on

disability type, indicating specific disabilities influenced vaccination behavior. However,

demographic indicators such as race/ethnicity and age did not significantly correlate with

vaccination uptake.

Research Question 3 delved into the interaction effect between disability and

reasons for not receiving vaccinations and assessed whether sociodemographic indicators

moderated the relationship. The logistic regression analysis demonstrated significant

predictive power, with the model explaining a substantial proportion of the variance in the

reasons for not receiving vaccinations. The findings unveiled a nuanced relationship

between disability status and reported reasons for not receiving vaccinations, with

sociodemographic indicators moderating this association.

The findings from Chapter 4 underscored the complex interplay between disability

status, disability types, sociodemographic indicators, and COVID-19 vaccination uptake.

The lack of significant association between disability status and vaccination uptake, as

well as the nuanced differences observed among various disability types, revealed critical

insights into the inequities faced by individuals with disabilities. These insights were

crucial for developing targeted interventions and policies to mitigate these inequities and

promote health equity.

The study's results also highlighted the moderating role of sociodemographic

indicators such as race/ethnicity and age in shaping vaccination rates and reasons for

vaccine hesitancy. The findings suggested that public health strategies must consider

these intersecting factors to effectively address the unique barriers encountered by

different subgroups within the disabled community.

Understanding these dynamics was essential for informing public health

interventions and policy decisions. By translating these findings into actionable

recommendations, we could develop more inclusive and practical strategies to improve

vaccination rates and reduce health inequities among individuals with disabilities.

Chapter 5 is built on the findings presented in Chapter 4, moving from analysis to

action. In this chapter, I outlined specific recommendations, discussed the study's broader

implications, and provided concluding thoughts emphasizing the importance of

intersectional approaches in public health research and practice.

Chapter 5: Discussion, Conclusions, and Recommendations

Introduction

In Chapter 5, I discuss the implications of the study, including potential positive

social changes and future research and practice recommendations. In this chapter I also

synthesize the study's contributions to public health, emphasizing the importance of

intersectional approaches in addressing vaccination inequities and promoting health

equity among individuals with disabilities. I examined the factors affecting COVID-19

vaccination uptake among individuals with disabilities, emphasizing the intersection of

disability status, sociodemographic indicators like race, ethnicity, and age, as well as

vaccination status. I used a quantitative cross-sectional design to analyze vaccination

rates and the reasons behind vaccine hesitancy among adults aged 18 and older in the

United States. I conducted this study to address the observed inequities in vaccination

rates among people with disabilities, identify the sociodemographic indicators influencing

these inequities, and inform public health interventions to ensure equitable vaccine

access.

There was no significant association between disability status and COVID-19

vaccination uptake, and sociodemographic indicators such as race, ethnicity, and age,

along with their interactions with disability status, did not significantly predict

vaccination status. However, the logistic regression model revealed significant differences

in vaccination rates based on disability type, with specific types of disabilities influencing

vaccination uptake. Nonetheless, demographic indicators did not show a significant

correlation. Logistic regression analysis indicated that disability status significantly

predicted the reported reasons for not receiving vaccinations, and sociodemographic

indicators moderated the relationship between disability status and vaccination hesitancy.

The regression model showed a small proportion of the variance in vaccination uptake,

with an R² of .012, and significant predictors within the model highlighted the importance

of considering both disability type and sociodemographic indicators to understand

vaccination behavior.

Several complex factors, including chronic health conditions, misinformation, and

accessibility issues, shape the current landscape of vaccine hesitancy among individuals

with disabilities. The article by Hinson-Enslin and Espinoza (2024) highlighted that,

individuals with sensory disabilities, particularly those with mental health conditions,

exhibited higher rates of anxiety, depression, and vaccine hesitancy, often due to distrust

in the vaccine and the government, necessitating tailored communication strategies to

address these concerns. Similarly, the scoping review by Nkambule and Mbakaya (2024)

identified myths and misinformation spread via social media and religious leaders as

significant factors contributing to vaccine hesitancy in Malawi, suggesting the importance

of targeted communication to counter these misconceptions. Furthermore, Jessica

Dimka's (2024) study revealed that people with chronic health conditions in Oslo were

more likely to accept COVID-19 vaccines compared to those without such conditions,

while individuals with disabilities faced more significant challenges in accessing

vaccines, underscoring the need for improved public health communication and

accessibility to ensure equitable vaccine distribution and uptake among vulnerable

populations. According to Charles et al. (2024), the future of the adult vaccine landscape

is rapidly evolving due to scientific and technological advancements and an increased

focus on the societal and economic benefits of vaccines.

Interpretation of the Findings

In this study, I confirmed previous findings that individuals with intersecting

marginalized identities experienced compounded forms of disadvantage. The lack of

significant association between disability status and vaccination uptake aligned with

Breaux and Rooks (2022), who found that race/ethnicity and disability interacted to

influence flu vaccine uptake. I extended this understanding to COVID-19 vaccination,

highlighting how specific types of disabilities affected vaccination rates. Moreover, the

findings aligned with Clemente et al. (2022) and Gréaux et al. (2023) regarding the

significant barriers individuals with disabilities face in accessing healthcare services,

including vaccination. These barriers were compounded by intersecting

sociodemographic factors such as race/ethnicity and socioeconomic status.

Contrary to some studies (e.g., Javed et al., 2022), I found no significant

correlation between sociodemographic indicators such as race/ethnicity and age and

COVID-19 vaccination uptake. This divergence suggested that while sociodemographic

factors were critical, their impact on vaccination uptake might vary depending on the

specific context and population studied. In this study, I challenged the effectiveness of

some existing public health interventions by highlighting that many such initiatives failed

to address the cultural and social contexts of intersecting identities. This aligned with

Marfo et al. (2024), who emphasized the need for culturally tailored interventions to

address historical and contemporary barriers to vaccine access.

I extended existing knowledge by revealing that specific types of disabilities

significantly influenced vaccination uptake. This added depth to understanding how

different disabilities intersected with sociodemographic factors to affect health behaviors,

which was less explored in previous studies. By employing an intersectionality

framework, I developed a more holistic view of how multiple social identities and

systems of oppression interacted to influence health outcomes. I extended the work of

Harari and Lee (2021) by providing empirical evidence on the complexities of

intersectional health inequities, particularly in the context of vaccination uptake. The

logistic regression model showed that disability status significantly predicted the reasons

for not receiving vaccinations, with sociodemographic indicators moderating this

relationship. This finding extended the literature by demonstrating the nuanced interplay

between disability, sociodemographic factors, and vaccine hesitancy, aligning with the

theoretical frameworks of intersectionality and social determinants of health.

The study's findings underscored the importance of intersectionality in

understanding health inequities. By highlighting the compounded disadvantage

experienced by individuals with intersecting marginalized identities, the findings in this

study confirmed Crenshaw's (1989) assertion that social identities intersected to produce

unique experiences of oppression and privilege. The findings revealed the complexity of

health behaviors and outcomes, moving beyond single-axis approaches focused on one

identity dimension. This complexity was crucial for developing more inclusive public

health strategies that addressed the specific needs of diverse populations. I addressed the

challenges of operationalizing intersectionality by using a logistic regression model to

analyze how different disability types and sociodemographic indicators interacted to

influence vaccination uptake. This methodological approach offered a practical example

of how intersectionality could be quantified in empirical research.

The findings emphasized the role of sociodemographic factors in shaping health

outcomes, consistent with the social determinants of health framework. The lack of a

significant correlation between sociodemographic indicators and vaccination uptake

suggested that these determinants interacted in complex ways that required further

exploration. The research highlighted the significant barriers individuals with disabilities

face in accessing healthcare services, reinforcing the importance of addressing social

determinants such as socioeconomic status, race/ethnicity, and disability in public health

interventions. The findings suggested that public health interventions must consider the

intersecting factors that shaped health behaviors and outcomes, aligning with the social

determinants of health framework, which advocated for addressing the broader social and

environmental factors contributing to health inequities.

Limitations of the Study

One fundamental limitation affecting the generalizability of the study was its

reliance on self-reported data from the HPS. While the HPS was a national survey, the

inherent biases of self-reported data, such as social desirability and recall bias, could have

impacted the accuracy and reliability of the findings (Rosenman et al., 2011).

Additionally, the study focused on adults aged 18 and older in the United States,

excluding individuals under 18. This exclusion limited the applicability of the results to

the broader population, particularly children and adolescents with disabilities who might

have experienced different healthcare challenges and outcomes.

The trustworthiness of the study's findings was constrained by the potential biases

associated with the online survey format of the HPS (Oliveri et al., 2021). Digital access

and literacy issues might have led to the underrepresentation of specific subgroups within

the disability community, such as those with limited internet access or lower digital

literacy. This underrepresentation could have skewed the results, making them less

reflective of the disabled population. Furthermore, the study's cross-sectional design only

provided a snapshot of the data at a specific point in time, limiting the ability to draw

causal inferences or observe changes over time (Capili, 2021).

The study's internal validity was influenced by the quality of the self-reported

data, which may not have always accurately reflected individuals' actual disability status,

sociodemographic characteristics, and healthcare experiences. Reporting biases, such as

over- or under-reporting of vaccination status and healthcare access issues, could have

affected the validity of the findings (Stephenson et al., 2022). Additionally, I could not

validate the self-reported data against external sources, further impacting its internal

validity. The complexity of measuring intersectionality through logistic regression models

might have also introduced challenges in accurately capturing the nuanced interplay of

multiple social identities and their compounded effects on health outcomes (Levandowski

et al., 2024).

Reliability issues arose from the study's reliance on a single HPS dataset, which

may not have consistently captured all relevant variables over time. The dynamic nature

of the COVID-19 pandemic and changing public health policies could have led to

variations in survey responses, impacting the consistency and repeatability of the

findings. Additionally, self-reported measures for critical variables, such as vaccination

uptake and reasons for vaccine hesitancy, might have been subject to individual

perceptions and reporting accuracy fluctuations, further affecting the study's reliability

Recommendations

It is essential to include individuals under 18 to expand the demographic scope of

future studies. By focusing on children and adolescents with disabilities, researchers can

comprehensively understand healthcare challenges and outcomes across all age groups.

This inclusion is crucial as younger individuals with disabilities may face unique issues

that are not adequately represented in studies limited to adults. Addressing the needs of

this younger population can lead to more targeted and effective healthcare interventions.

Broadening the representation of disability subgroups is another vital

recommendation. Efforts should be made to include diverse subgroups within the

disability community, particularly those with limited digital access or lower digital

literacy. Ensuring these groups' inclusion will result in more representative and

generalizable findings. This broader representation will result in the identification of the

specific needs and challenges various subgroups face, leading to more inclusive

healthcare policies and practices.

Enhancing data collection methods can significantly improve the quality of

research. Combining self-reported data with objective measures and qualitative

interviews will provide a more nuanced and validated understanding of the experiences of

individuals with disabilities. A mixed-methods approach can uncover deeper insights and

ensure the reliability of findings. Additionally, implementing a longitudinal study design

will allow researchers to observe changes over time and better assess causal relationships

between variables. This design is crucial for understanding the long-term impacts of

healthcare interventions and policies.

Addressing reporting biases is another critical area for improvement. Future

research should validate self-reported data against external sources, such as medical

records or third-party surveys, to enhance accuracy and reliability. Refining survey

instruments to minimize social desirability and recall biases is also necessary. This can be

achieved using more precise and neutral wording and techniques such as diaries or

timeuse surveys to aid recall. These improvements will lead to more accurate data and

more reliable conclusions.

Examining intersectionality in greater depth is essential for capturing the complex

interplay of multiple social identities and their compounded effects on health outcomes.

Employing advanced analytical techniques, such as intersectional mixed-effects models,

can help achieve this goal. Additionally, conducting subgroup analyses that address the

intersectionality of different demographic factors, such as age, gender, race, and

socioeconomic status, will provide deeper insights into the varied experiences within the

disability community. This approach will help in developing more tailored and effective

healthcare strategies.

Improving the reliability of research through the use of diverse data sources is

another crucial recommendation. Using multiple datasets from different sources can result

in triangulated findings and enhance the consistency of the research. Regularly updating

the data collection process and conducting follow-up studies will account for the dynamic

nature of public health situations, such as the COVID-19 pandemic. This approach may

result in more up-to-date and relevant findings, ensuring that research remains pertinent

and actionable.

Implications

Positive Social Change

At the individual level, understanding the health inequities faced by individuals

with disabilities can result in positive social change by leading to better-targeted

healthcare interventions. Improved health outcomes can be achieved through personalized

care plans that address the unique needs of individuals with various types of disabilities,

incorporating their specific health conditions, sociodemographic indicators, and

vaccination status. Empowering individuals with disabilities and their caregivers with

evidence-based information about their health challenges can foster advocacy efforts for

better services and accommodations, promoting a more equitable healthcare landscape.

At the family level, more profound insights into health inequities can strengthen

support systems within families of individuals with disabilities. Family education

programs can help caregivers navigate healthcare systems, access resources, and provide

adequate care for their disabled family members. The study supports the development of

family-centered care models that integrate the needs and roles of family members in

managing the health and well-being of individuals with disabilities, fostering a

collaborative approach to healthcare decision-making.

At the organizational level, the study’s findings can drive positive social change

by promoting inclusive policies and practices that accommodate the needs of individuals

with disabilities. Organizations, including healthcare providers and employers, can

develop more inclusive policies that ensure equitable access to services and

accommodations. Training programs for healthcare providers on disability competence,

cultural competence, and intersectionality can enhance the quality of care provided to

individuals with disabilities, fostering a more supportive and inclusive healthcare

environment.

At the societal and policy level, policymakers can use the study’s findings to

develop policies that address the systemic barriers faced by individuals with disabilities.

By improving accessibility to healthcare services and ensuring equitable vaccine

distribution, policymakers can work towards reducing health inequities and promoting

health equity across the broader population. Incorporating intersectionality into health

policy development is crucial to address the compounded effects of multiple marginalized

identities, ensuring inclusive and equitable policies. Governments should allocate funding

to initiatives to improve the accessibility of healthcare facilities and services for people

with disabilities, fostering a more inclusive and accessible healthcare system for all.

Methodological, Theoretical, and Empirical Implications

The methodological, theoretical, and empirical implications of these findings

suggest the importance of considering diverse perspectives and factors when conducting

public health and healthcare research. By incorporating variables such as disability status

and types, and sociodemographic factors into analyses, researchers can gain a more

comprehensive understanding of health disparities and the effectiveness of interventions.

This can lead to the development of more accurate models and strategies for promoting

health equity and improving healthcare outcomes for marginalized populations.

Recommendations for Practice

Practice recommendations include the implementation of targeted interventions

and outreach efforts to increase vaccination uptake among individuals with disabilities.

This may involve creating accessible vaccination sites, providing educational materials in

multiple formats, and offering support services to address specific barriers this population

faces. Moreover, healthcare professionals should receive training on how to effectively

communicate with and support individuals with disabilities to ensure they have equal

access to healthcare services. By incorporating these recommendations into practice,

healthcare organizations can work towards reducing health disparities and promoting

positive social change in their communities.

Conclusion

This study illuminated the intricate web of health inequities that individuals with

disabilities face, particularly in the context of COVID-19 vaccination uptake during the

pandemic in the US. While inequities persisted, understanding and addressing the

multifaceted barriers faced by this population were essential for achieving health equity.

By delving into the intersectionality of disability status, sociodemographic indicators, and

vaccination status, it underscored the urgent need for more inclusive healthcare

approaches. The findings emphasized the imperative of tailored interventions, inclusive

policies, and proactive support systems to bridge the gap in healthcare access and

outcomes. Ultimately, this study served as a clarion call for society to embrace diversity,

promote equity, and ensure that no one is left behind in pursuing health and well-being.

In this study, I examined the intersectional impacts of disability status,

sociodemographic indicators (e.g., race/ethnicity, socioeconomic status), and geographic

location on COVID-19 outcomes (infection rates, hospitalizations, mortality) and access

to healthcare services (testing, treatment, vaccination) for individuals with different types

of disabilities in the United States during the COVID-19 pandemic. Individuals with

disabilities face unique challenges and barriers in accessing healthcare services and public