University of Southern California Dissertations and Theses

Relationship between L.A. County residents' demographics and willingness to take the COVID-19 vaccine

(USC Thesis Other)

Relationship between L.A. County residents' demographics and willingness to take the COVID-19 vaccine

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content RELATIONSHIP BETWEEN L.A. COUNTY RESIDENTS’ DEMOGRAPHICS
AND WILLINGNESS TO TAKE THE COVID-19 VACCINE

by

Nicole Karpowicz

A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
BIOSTATISTICS

December 2021

Copyright 2021 Nicole Karpowicz
ii
TABLE OF CONTENTS

DEDICATION iii
ACKNOWLEDGEMENTS iv
ABSTRACT v
INTRODUCTION 1
METHODS 8
STATISTICAL ANALYSIS 11
RESULTS 13
DISCUSSION 15
CONCLUSION 18
REFERENCES 19
TABLES 21
FIGURES 22

iii
DEDICATION

This thesis is dedicated to my unconditionally loving parents. They have been a constant
source of inspiration and support throughout my life and have done everything possible to help
me get to where I am today.

iv
ACKNOWLEDGEMENTS

I would like to give special thanks to my thesis committee chair, Dr. Kayla de la Haye,
for giving me guidance and expertise during the thesis development process. Additional thanks
for letting me join in on her research endeavor with the Stay Connected L.A. team, which gave
me additional insight into how the pandemic disparately affects different groups of people and
inspired this thesis.
I would also like to give thanks to another thesis committee member, Dr. Lourdes Baez-
Conde for sharing all her knowledge and expertise and further stimulating my interest in public
health disparities in Los Angeles.
Lastly, I would like to thank my advisor and thesis committee member, Dr. Wendy Mack.
Thank you for all the advice and guidance you have given me while pursuing my Biostatistics
Master’s degree.

v
ABSTRACT
Background: When COVID-19 first emerged in December of 2019, the world radically changed
to combat the novel virus. Society had to adapt to a new normal that consisted of staying home,
socially distancing, hand-washing frequently, and wearing masks in public. The only way to
bring an end to the pandemic and return to regular life is to reach herd immunity- when enough
of the population is immune that the virus stops spreading. However, some barriers to herd
immunity exist in the form of vaccine hesitancy or unwillingness to take the vaccine. This thesis
focused on associations between demographic features of Los Angeles County residents and
vaccine hesitancy.
Methods: The data used for this analysis come from the Understanding Coronavirus in America
Study. This study is a longitudinal panel survey that over time follows about 1800 L.A. County
adults and gathers information on a variety of topics, including Covid-19 vaccine hesitancy. The
panel asked specific questions about whether or not participants intended to take the vaccine. We
then analyzed by demographics including age, gender, household income level, and
race/ethnicity. We tested for statistically significant associations between these demographic
variables and vaccine hesitancy using Pearson Chi-Square Tests. Logistic regression was used
for ordered variables (age and household income level) to test for a directional relationship
amongst these variables and vaccine hesitancy.
Results: All demographic variables analyzed had statistically significant relationships with
vaccine hesitancy. Women were significantly more vaccine hesitant than men (p<0.001) with
only 64% of women saying they were likely to get the vaccine, compared to 76.2% of men. Age
was also a significant determinant of vaccine hesitancy (p<0.001) with older individuals being
more likely to vaccinate than younger ones. Different household income levels were also a factor
vi
that affected willingness to get the vaccine (p<0.001) with lower income levels being less likely
to get the vaccine than higher ones. Race/ethnicity was also a statistically significant feature
associated with vaccine hesitancy (p<0.001) as non-Hispanic Whites and Asians were more
likely to get the vaccine while Hispanic/Latino and Black/African American populations were
more likely to be hesitant. We performed a logistic regression on vaccine hesitancy with age and
found that they are linearly related, as the older someone is, the more likely they are to get the
vaccine. Logistic regression on vaccine hesitancy and household income level showed that they
linearly correlate as when income increases, so does willingness to get the vaccine.
Conclusion: This analysis helps to identify the characteristics of Los Angeles County adult
residents who are the most hesitant to get the COVID-19 vaccine. In this sample, younger adults,
women, people with low household incomes, and Hispanics/Latinos and Blacks/African
Americans were all more likely to be hesitant about getting the vaccine. Going forward,
specifically targeting these groups with tailored vaccine campaigns could diminish the number of
vaccine hesitant individuals in the population.

1
INTRODUCTION
COVID-19 Background
Our world radically changed with the emergence of the new viral pathogen, SARS-CoV-
2, now commonly referred to as COVID-19. In December of 2019, the novel virus was
discovered in the Hubei province of China, when an outbreak of pneumonia suddenly affected
many people. It is speculated that the pathogen is from animal origin, and eventually made the
jump to humans at the Huanan market in Wuhan (Amawi, Abu, Aljabali, Dua & Tambuwala,
2020). Once humans got infected, the virus rapidly spread and infected many others in a short
amount of time.
SARS-CoV-2 is a member of the coronavirus family of viruses. This family also
contains the respiratory viruses that caused the outbreaks of SARS in China and MERS in the
Middle East (Amawi et al, 2020). Coronaviruses are named for their spiked appearance which
results from their protein structure. The SARS-CoV-2 genome encodes its unique shape with
four main structural proteins: the nucleocapsid protein, the membrane protein, the spike protein,
and the envelope protein. The S glycoprotein (spike protein) is located on the surface of the virus
and interacts with host cell surface receptors to infect and take over the host cell’s machinery
(Boopathi, Poma & Kolandaivel, 2020). Once the virus takes over the host cell, it can replicate
and spread throughout the infected person.
The virus that causes COVID-19 attacks the lower respiratory system, potentially leading
to severe disease. Current evidence shows that the disease is primarily spread through the
inhalation of infected droplets, such as when a coronavirus patient coughs (Uddin et al, 2020).
An important indicator of viral spread at the population level is the R0 value. This value is used
in epidemic models for disease transmission, like the SIR model, and represents the reproduction
2
number, informing us of the average number of people that will get infected (in the population of
susceptible individuals) for each person that contracts the disease. The R0 for COVID-19 is
estimated to be between 2 and 3, indicating that when one person contracts the novel
coronavirus, two to three more people will subsequently get infected. This R0 value is higher
than that of the seasonal flu (Brussow, 2020). The reproduction number could possibly change
with new mutations in the virus, leading to more infectious strains.
COVID-19 was first discovered due to a sudden onslaught of pneumonia cases in China.
Since then, we have come to learn that there is a broad spectrum of symptoms that are associated
with the disease. These can range from asymptomatic infections to mild forms to severe cases
that may result in death. Some risk factors for contracting a more severe form of the disease, and
death, include older age (65+) and underlying health conditions such as obesity, diabetes, and
autoimmune disorders. Common symptoms of the disease include a dry cough, shortness of
breath, fever, as well as some gastrointestinal symptoms. (Uddin et al, 2020). According to the
New York Times, as of January 2021, over 100 million cases have been reported worldwide with
over 3 million deaths. Infection with COVID-19 is diagnosed using a real-time polymerase
chain reaction (RT-PCR), which takes a nucleic acid sample from the subject and then amplifies
the sequence. RT-PCR has shown to be both sensitive and specific in detecting the presence of
the virus. It uses specific primers and probes to get a fairly rapid result as to whether or not
people have SARS-CoV-2 in their systems at the time of the test (Ahn et al, 2020).
Pandemic Spread
When the novel coronavirus emerged in China in December of 2019, the country had to
rapidly respond to the outbreak. On January 23, 2020, Chinese authorities closed off the entire
city of Wuhan as a measure of containing the virus. All transportation into and out of the city
3
was cut off and individuals were required to quarantine. However, the virus still managed to
spread out of the region as travelers that were visiting the city had brought the virus back to their
home countries. Cases started to appear in Europe and in the United States with large waves
hitting different countries at varied times. Italy was one of the pandemic’s first hard-hit countries
with high hospitalization rates and widespread lockdowns. The virus continued on to devastate
more countries around the globe. On January 30
th
, the World Health Organization had declared
the virus a global health emergency and by March 11
th
the virus was declared a pandemic
(Cucinotta & Vanelli, 2020).
Public Health Efforts to Contain the Virus
Many measures were taken in an effort to curb the disease’s escalation. Some countries
closed their borders to travelers (e.g., China, Australia, New Zealand, Taiwan) while others kept
their borders relatively open (e.g., U.S.). Additionally, government leaders instituted various
degrees of stay at home orders in which non-essential businesses shuttered and individuals were
required to quarantine themselves in their homes. As places started to reopen again, protective
measures such as six feet social distancing, mask-wearing, and frequent hand washing became a
part of what is referred to as the “new normal” (Chu et al, 2020). Nevertheless, although these
measures are essential for slowing viral spread and preventing health systems from getting
overwhelmed, they will not be sufficient to bring an end to the pandemic.
Pandemics end when enough of the population is immune to the disease due to prior
infection, natural immunity, or acquired immunity by vaccine. When this occurs, the population
is said to have gained herd immunity. Herd immunity is a concept in which an adequate
proportion of the population is immune to the disease which will prevent the further propagation
of the disease - even with certain individuals in the population not having immunity. Herd
4
immunity is directly related to the R0 of a virus. More infectious illnesses will require a higher
level of immune individuals in order for everyone to be protected. Since SARS-CoV-2 is
estimated to have an R0 value between 2 and 3, one estimate suggests that approximately 60% of
the population would need to be immune in order to provide herd immunity (Brussow, 2020).
One possible end to the pandemic is to reach herd immunity by allowing the virus to remain on
its natural course and infect as many people as possible. However, this method comes at too
great a cost as many people will unnecessarily die. The preferable and life-saving alternative is to
vaccinate the population to the required level in order to reach herd immunity.
Vaccine Development
Vaccines provide people with immunity to diseases by introducing the pathogen in an
innocuous form. The immune system learns to recognize the invader by developing antibodies
that will be prepared to attack the virus when the real version comes. According to the FDA, a
vaccine needs to undergo several scheduled stages of testing in order for the vaccine to be
authorized for public use. The first step involves research and development in which the actual
vaccine is constructed. Next comes the pre-clinical stage where the vaccine is tested on animals
to determine if it is safe and produces the desired reaction. Then comes four stages of clinical
testing on human subjects. Phase 1 consists of a small group of participants (<100 people) and
examines the initial safety and dosage of the vaccine. Phase 2 consists of a larger group of
participants (<1000 people) and continues to check the safety and side effects, as well as the
early effectiveness of the vaccine. Phase 3 further expands to thousands of people to check
longer term safety and effectiveness among many different groups of people. Finally, after
undergoing all these stages, the vaccine can get submitted for FDA approval. Following
5
approval, phase 4 will examine the long-term safety and effectiveness of the vaccine in a large
population.
The development and FDA approval of a vaccine is a long process, with a normal
timeline of about 15-20 years. However, with the COVID-19 pandemic taking lives every day,
this process was accelerated to grant a COVID-19 vaccine emergency authorization in 1-1.5
years. In order to achieve this rapid timetable, the U.S. Department of Health and Human
Services launched the vaccine initiative known as Operation Warp Speed. Participating
companies received funding from the federal government to fast track the development of a
COVID-19 vaccine. The goal of this partnership between the government and private companies
was to deliver 300 million safe and effective doses of the vaccine to the public by the end of
January, 2021 (O’Callaghan, Blatz & Offit, 2020).
There are currently multiple vaccines in the final stages of trials and a few that have even
been given FDA emergency use authorization in the United States. The leading two vaccines in
the U.S. that have been granted early approval are the Pfizer and Moderna vaccines with Pfizer
reporting a 95% efficacy and Moderna reporting a 94.5% efficacy. Both shots were created with
a novel technology known as an mRNA vaccine. This technology inserts the genetic information
necessary to produce antibodies against the virus instead of a disabled form of the virus. These
vaccines tell cells in the body to build spike protein seen on SARS-CoV-2, in order to teach the
body to fight off the coronavirus (O’Callaghan et al, 2020).
Vaccine Hesitancy
Unfortunately, developing a safe and effective vaccine is only half the battle. We also
need people to take it. Vaccine hesitancy threatens the ending of the pandemic as it will be more
difficult to reach the threshold necessary for herd immunity. It is estimated that 5-10% of
6
individuals in the developed world have strong negative opinions about vaccines in general
(Dube et al, 2013). This number could be higher for vaccines that are quickly developed during
an ongoing health crisis, such as the one for COVID-19.
Vaccine hesitancy has been a long-standing obstacle with the development of new
vaccines. The World Health Organization has extensively studied hesitancy among influenza
vaccines. They studied many reasons for hesitancy including illness risk perception, social
pressures, cues to action, and others. In general, when people did not perceive the disease in
question as risky, they were less likely to vaccinate. Social pressure was found to positively
impact vaccine intent as when society establishes vaccinating as the norm, people were more
likely to get a vaccine. Also, cues to action such as receiving recommendations from medical
personnel positively impacted a person’s willingness to get a vaccine (Schmid, Rauber, Betsch,
Lidolt & Denker, 2017).
Much of vaccine hesitancy is also dependent on socio-cultural context. Prior negative
experiences with health services or living in a community of vaccine-hesitant individuals could
influence people to become or remain vaccine hesitant. For example, many Latinos and African
Americans have a general distrust of medicine due to individuals experiencing healthcare
discrimination. Over time, this distrust further progresses to more and more people throughout
these communities. Additionally, through the media and the introduction of politics into public
health discussions, there seems to be an increase in mistrust and people may create narratives
about ulterior motives for a vaccine. This reason is especially likely in the currently highly
charged political context in the United States. Another possible factor in vaccine hesitancy is
perceived risk from vaccination. Some individuals may worry that the vaccine could cause the
illness itself or other adverse events. This fear could be escalated particularly with a vaccine that
7
was developed in a much shorter time period than is traditional. Although the purpose of a
vaccine is to lower the risk of contracting a disease, people are more averse to risk associated
with an action (such as taking a vaccine) than they are to risk associated with an inaction
(contracting COVID-19). This omission bias can lead to people preferring to deal with the risk of
the virus over the risk of actively taking a vaccine (Dube et al, 2013).
Considering the important role vaccines play in a population’s reaching herd immunity
and the state of the ongoing pandemic, this thesis focused on vaccine hesitancy within
individuals in Los Angeles County. We investigated the overall willingness to take the vaccine
within this population. We delved deeper into analysis and stratified the population by multiple
demographic variables to compare vaccine hesitancy Establishing which portions of a population
appear to have greater levels of vaccine hesitancy is the first step in understanding these beliefs
and taking further action to reduce them.

8
METHODS
Study and Participants
The data that were analyzed for this thesis came from the Understanding Coronavirus in
America Study, which is embedded in a larger longitudinal panel study known as the
Understanding America Study (UAS). Participants were sampled based off of zip codes from a
commercial vendors’ post office delivery sequence. UAS is an internet panel survey by the
University of Southern California in which participants respond to survey questions. Participants
received compensation of $20 for every 30 minutes they spent working on a survey. UAS
continually collects data from around 9000 households across the U.S. and follows them for an
extended period of time. For the purposes of this paper, data analysis was focused only on the
approximately 1800 Los Angeles County participants in the panel.
UAS sampled participants in a manner meant to be representative of L.A. County
households, but additional measures were taken in the analysis to ensure accurate portrayal of the
population through the use of survey weights. Survey weights are a fractional value given to each
participant that determines how much influence the data point should have on the overall
analysis. It reflects the probability of survey sampling and allows us to properly estimate
population parameters. For example, survey responses from over-sampled populations would
have lower survey weights than from those of under-sampled populations. The survey weights
were part of the UAS dataset and were labeled with the variable final_weight. They were
developed specifically for the L.A. County subsample and for each particular wave. The weights
were used in all subsequent analyses.
For this analysis, we focused on the survey responses from one survey wave: The UAS
264 survey that was fielded from September 30
th
to October 27
th
, 2020. Preliminary analyses
9
(described below) indicated that this one wave of data was representative of vaccine intentions
from prior months’ surveys, which did not significantly differ.
Measures
Dependent variables. The outcome of interest was COVID-19 vaccine intentions among
L.A. County residents. Two variables in the datasets assessed vaccine intentions, the dependent
variables used in the analysis. The first dependent (variable cr030), was assessed by the question
“How likely are you to get vaccinated for coronavirus once a vaccination is available to the
public” (1=very unlikely, 2= somewhat unlikely, 3=somewhat likely, 4=very likely, 5=unsure).
Unsure responses were excluded from all analyses. The second dependent variable was vc005a,
which asked study participants whether “I would get the vaccine once it becomes available to
me” (1=yes, 2=no). We again made sure to remove any missing data so our analysis would not
be affected.
Explanatory variables. To understand how COVID-19 vaccine intentions differ among the L.A.
County residents with different demographics, we focused on differences based on gender, age
group, race and ethnic group, and income levels. These demographic characteristics are the
independent variables (or explanatory variables) for the analysis.
Participants had to specify their gender as either female or male (0=female, 1=male).
Their age was reported as date of birth, and categorized into age groups categories that we
recoded as “agecat”: 18-30 years, 31-40 years, 41-50 years, 51-64 years, and 65+ years.
Ethnicity was measured by asking participants if they identified as Hispanic or Latino and race
was determined by asking which racial group they belong to. These two measures were used to
compute the following race and ethnic categories: non-Hispanic White, Hispanic/Latino (for
those that identify as Hispanic of Latino, of any race), Asian (non-Hispanic), or Black (non-
10
Hispanic). The number of respondents in the other race and ethnicity categories were small, and
so were excluded from our analyses.
Participants’ household income was assessed by the question “Which category represents
the total combined income of all members of your family”, and reported as one of the following
categories: Less than $5,000 (encoded as 1), $5,000 to $7,499 (encoded as 2), $7,500 to $9,999
(encoded as 3), $10,000 to $12,499 (encoded as 4), $12,500 to $14,999 (encoded as 5), $15,000
to $19,999 (encoded as 6), $20,000 to $24,999 (encoded as 7), $25,000 to $29,999 (encoded as
8), $30,000 to $34,999 (encoded as 9), $35,000 to $39,999 (encoded as 10), $40,000 to $49,999
(encoded as 11), $50,000 to $59,999 (encoded as 12), $60,000 to $74,999 (encoded as 13),
$75,000 to $99,999 (encoded as 14), $100,000 to $149,999 (encoded as 15), and $150,000 or
more (encoded as 16). Any missing data was removed from the analysis. We recoded this data
into the following simplified categorical variable (“income”) for analysis: $0-$29,999 a year,
$30,000-$59,999 a year, $60,000-$99,999 a year, and $100,000+ a year.

11
STATISTICAL ANALYSIS
All statistical analyses for this thesis were performed using the University Edition of the
Statistical Analytical Software (SAS) software. To ensure that the data did not significantly
change over time, we ran an initial analysis that read in the datasets for UAS 246 and UAS 264.
They were then saved into SAS internal data files, mydata and mydata2. The data files include
all UAS participants, and so we first subset data files to exclude any participants that did not live
in LA County. Our cleaned data for each survey was saved into new data files, newdata and
newdata2, and then we proceeded with the analysis.
The preliminary analysis sought to determine whether analyzing just one wave of data
reasonably encompassed vaccine intentions throughout the course of the pandemic (up until
October, 2020). Analyzing the datasets UAS 246, which collected information from May 27,
2020 to June 23, 2020, and UAS 264, which collected data from September 30, 2020 to October
27, 2020, we determined that vaccine intentions and hesitancy did not significantly change over
the course of the few months. For the month of June, the average value for likelihood of taking
the vaccine once it becomes available was 3.32 (1= very unlikely, 4 = very likely). In October,
the average value was 3.23 (1= very unlikely, 4 = very likely). We used proc surveymeans in
SAS to estimate these averages. Next, in order to see whether the means were significantly
different, we used the SAS function proc ttest. This allowed us to perform a two-sided, two-
sample t-test with an alpha value of 0.05. The null hypothesis was that the two means are equal
to one another, while the alternative hypothesis was that they are unequal. The test gave a p-
value of 0.057, and therefore fails to reject the null hypothesis (Table 1, Figure 1). There was not
a statistically significant difference in vaccine intentions between the months of June and
12
October, so it is reasonable to perform the rest of our analyses using the October survey wave,
instead of focusing on change in intentions over time.
Our statistical analysis focused on vaccine intentions among L.A. County residents,
estimating and testing for differences in these intentions based on participant demographics. All
data for demographic analysis were taken from the UAS 264 (September 30 to October 27, 2020)
survey. For this analysis, we used the vc005a variable which asked whether or not participants
would get the vaccine when it became available to them (1=yes, 2=no). We computed summary
statistics for all L.A. County residents and then tested if the proportions that would take the
vaccine differed for gender, age, race/ethnicity, and household income. We used the Pearson
Chi-Square test for proportion comparison to test if vaccine intentions were significantly
different. For ordinal variables (age and income) that had a statistically significant difference
among groups we ran a logistic regression, using the SAS function proc logistic, to test if there
was a directional relationship between willingness to take the vaccine and the ordered groups.

13
RESULTS
Overall, 68.5% (95% CI 65.7-71.3) of all L.A. County residents indicated they would get
the COVID-19 vaccine if it became available to them. Stratifying by gender, 75.9% (95% CI
71.7-80.0) of men vs. 63.6% (95% CI 59.8-67.4) of women said they would get the vaccine (p-
value <0.001). Women were significantly less likely to intend to get the vaccine. Stratifying by
age category, the proportion of each age group that intended to get the vaccine was 60.7% (95%
CI 54.0- 67.4) of 18 to 30 year olds, 64.8% (95% CI 58.7-71.0) of 31 to 40 year olds, 63.8%
(95% CI 56.9-70.8) of 41 to 50 year olds, 76.9% (95% CI 71.5-82.3) of 51 to 64 year olds, and
76.9% (95% CI 70.3-83.5) of adults 65 years and older (p-value <0.001). This indicates that
there is a significant difference between intention to take the vaccine between different age
categories in L.A. County.
Stratifying by household income, the proportion of each income group that intended to
get the vaccine was: 59.7% (95% CI 53.9-65.6) of adults with $0-29,999 yearly income, 62.3%
(95% CI 56.3-68.2) of adults with $30-59,999 yearly income, 70.9% (95% CI 65.0-76.9) of
adults with $60-99,999 yearly income, and 81.1% (95% CI 76.4-85.8) of adults with yearly
incomes of $100,000 and over (p-value <0.001). So, there is a statistically significant difference
between intention to take the COVID-19 vaccine across different income levels.
Stratifying by race/ethnicity, 77.7% (95% CI 73.5-82.0) of non-Hispanic Whites, 60.8%
(95% CI 56.1-65.6) of Hispanics/Latinos, 47.8% (95% CI 37.4-58.2%) of Blacks/African
Americans, 73.8% (95% CI 67.4-80.2) of Asians intend to take the vaccine when it becomes
available to them (p-value <0.001). So, there is a statistically significant difference in intent to
take the COVID-19 vaccine for the different race and ethnic groups. The proportions and p-
values from the analysis are summarized in Table 2.
14
After detecting statistically significant differences of willingness to take the vaccine
among every variable we stratified by, we further analyzed the ordinal variables to see how they
correlate with vaccine intent. Through logistic regression, we modeled the probability of
individuals getting the COVID-19 vaccine against their age. Our plot of the data and regression
model appeared fairly linear (Figure 2) and gave us a p-value of <0.001, indicating that there is a
statistically significant relationship between increasing age and greater willingness to take the
vaccine. The linear relationship shows that as age increases, so does the probability of getting the
coronavirus vaccine (Figure 2). Next, we modeled the probability of people willing to get the
vaccine against their household income, with group 1 containing the lowest yearly earnings and
group 16 containing the highest yearly earnings. Our logistic regression model appeared linear
(Figure 3) with a p-value of <0.001, indicating a statistically significant relationship between
income and willingness to take the vaccine. As income increases, so does the probability of
intending to get the vaccine. A graph of this association is shown in Figure 3.

15
DISCUSSION
Our analysis of vaccine intentions within L.A. County shows which parts of the
population were less willing to get the coronavirus vaccine from September 30, 2020 to October
27, 2020. Women were significantly less willing to get a vaccine than men. Possible reasons for
this hesitancy could be false narratives that are circulating about the vaccine being unsafe for
pregnant women and causing infertility. Spread of such misinformation could be sparking fears
of the vaccine among the female population, engendering further hesitancy. Examining age
differences in vaccine intentions, younger adults were more hesitant to vaccinate than are older
adults. This discrepancy could be due to the fact that younger adults are generally less at risk of
contracting a severe form of COVID-19, so they may be less motivated to protect themselves and
others. Upon inspection of variation in vaccine hesitancy among different household income
levels, we found that people with lower incomes were less likely to want to vaccinate than
people with higher incomes. This could possibly be due to lower income households having less
resources to get access to a vaccine. For example, some households may not have internet
connection or personal computers to sign up for a vaccination appointment. Lastly, analyzing
vaccine hesitancy among different racial and ethnic groups showed that while desire to get the
vaccine was high among non-Hispanic White and Asian populations, fewer Hispanics/Latinos
intended to get the vaccine, and Blacks/African Americans were the least likely to intend to get
vaccinated. These vast differences among racial and ethnic groups could be due to healthcare
inequalities experienced by minority groups. The institutional racism that pervades throughout
healthcare could breed mistrust in these groups in the system that is meant to keep everyone
healthy.
16
Understanding which population sectors are vaccine hesitant and the possible reasons for
this hesitancy is highly significant information when trying to bring about the end to a pandemic.
Since herd immunity is the most feasible approach to manage the continual spread of disease,
vaccinating as many individuals as possible is the priority. Discovering which parts of the
population are vaccine hesitant allows further steps to be taken to try and diminish these
numbers. As women are less willing to get the vaccine than men, they should be targeted with
campaigns that dispel any misinformation and instill confidence in the vaccine. Younger age
groups, who are more vaccine hesitant, could be targeted through means such as social media to
convey the importance of vaccinations for the health of the community, not just the individual.
Lower household income populations could be targeted with new community initiatives that help
them gain easier access to vaccines. Also, public health officials could do some outreach to
Hispanic/Latino and Black/African American populations to try and build some trust in the
healthcare system and the vaccine. Creating targeted campaigns with goals of increasing vaccine
willingness amongst these particular groups would be meaningful next steps that come from
understanding the demographic differences in COVID-19 vaccine hesitancy.
The findings of this thesis have further application than just the current situation in the
world right now. As mentioned in the introduction, COVID-19 vaccine hesitancy extends beyond
the ongoing pandemic as the hesitant mindset pervades amidst all inoculations. Studies by the
World Health Organization have shown that vaccine behaviors for the seasonal flu are highly
similar to behaviors among pandemic influenza (Schmid et al, 2017). So, if someone does not
regularly get their yearly flu shot, then they are also less likely to get a vaccine during a global
health crisis. Also, it is estimated that up to 5-10% of people are generally against all vaccines
(Dube et al, 2013). Further in-depth studies have delved into the variety of reasons people give
17
for not wanting to get inoculated, which is important as our analysis of UAS data just provides a
quantitative snapshot at the situation. The WHO found psychological, physical, contextual, and
sociodemographic barriers can all impede someone from getting a vaccine (Schmid et al, 2017).
Combining knowledge from the WHO of why people don’t want to get the vaccine with this
thesis’ findings of who does not intend to vaccinate would greatly help in creating targeted
campaigns that aim at diminishing negative sentiments towards the COVID-19 vaccine as well
as vaccines in general.

18
CONCLUSION
This thesis examined vaccine intentions within different sectors of the community which
is highly relevant considering the current global pandemic. With the coronavirus contagion
sickening people around the world, every effort is being taken to bring about herd immunity.
Knowing exactly which subpopulations have greater opposition for the vaccine is useful
information when trying expand vaccination campaigns. Our analysis showed among L.A.
County residents, who number over 10 million, younger people, women, people with lower
household income, and Hispanic/Latino and Black/African American populations were the least
willing to get a vaccine. Public health officials can specifically target these groups with tailored
messaging and interventions to increase ease of access, to decrease hesitancy and promote
vaccine update. The findings from this thesis could also have more far reaching effects as
vaccine hesitancy seems to be consistent among all inoculations. So, decreasing COVID-19
vaccine hesitancy will also lead to decreasing hesitancy amongst all inoculations and overall
make the world a healthier place.

19
REFERENCES

Ahn, D. G., Shin, H. J., Kim, M. H., Lee, S., Kim, H. S., Myoung, J., Kim, B. T., & Kim, S. J.
(2020). Current Status of Epidemiology, Diagnosis, Therapeutics, and Vaccines for
Novel Coronavirus Disease 2019 (COVID-19). Journal of microbiology and
biotechnology, 30(3), 313–324. https://doi-
org.libproxy1.usc.edu/10.4014/jmb.2003.03011

Amawi, H., Abu Deiab, G. I., A Aljabali, A. A., Dua, K., & Tambuwala, M. M. (2020). COVID-
19 pandemic: an overview of epidemiology, pathogenesis, diagnostics and potential
vaccines and therapeutics. Therapeutic delivery, 11(4), 245–268. https://doi-
org.libproxy1.usc.edu/10.4155/tde-2020-0035

Boopathi, S., Poma, A. B., & Kolandaivel, P. (2020). Novel 2019 coronavirus structure,
mechanism of action, antiviral drug promises and rule out against its treatment. Journal
of biomolecular structure & dynamics, 1–10. Advance online publication.
https://doi.org/10.1080/07391102.2020.1758788

Brüssow H. (2020). Immunology of COVID-19. Environmental microbiology, 22(12), 4895–
4908. https://doi-org.libproxy2.usc.edu/10.1111/1462-2920.15302

Chu, D. K., Akl, E. A., Duda, S., Solo, K., Yaacoub, S., Schünemann, H. J., & COVID-19
Systematic Urgent Review Group Effort (SURGE) study authors (2020). Physical
distancing, face masks, and eye protection to prevent person-to-person transmission of
SARS-CoV-2 and COVID-19: a systematic review and meta-analysis. Lancet (London,
England), 395(10242), 1973–1987. https://doi-org.libproxy1.usc.edu/10.1016/S0140-
6736(20)31142-9

Commissioner, O. O. (n.d.). Step 3: Clinical Research. Retrieved from
https://www.fda.gov/patients/drug-development-process/step-3-clinical-research

Cucinotta, D., & Vanelli, M. (2020). WHO Declares COVID-19 a Pandemic. Acta bio-medica :
Atenei Parmensis, 91(1), 157–160. https://doi.org/10.23750/abm.v91i1.9397

Dube, E., Laberge, C., Guay, M., Bramadat, P., Roy, R., & Bettinger, J. (2013). Vaccine
hesitancy: an overview. Human vaccines & immunotherapeutics, 9(8), 1763–1773.
https://doi-org.libproxy2.usc.edu/10.4161/hv.24657

O'Callaghan, K. P., Blatz, A. M., & Offit, P. A. (2020). Developing a SARS-CoV-2 Vaccine at
Warp Speed. JAMA, 324(5), 437–438. https://doi.org/10.1001/jama.2020.12190

Schmid, P., Rauber, D., Betsch, C., Lidolt, G., & Denker, M. (2017). Barriers of Influenza
Vaccination Intention and Behavior – A Systematic Review of Influenza Vaccine
Hesitancy, 2005 – 2016. Plos One,12(1). doi:10.1371/journal.pone.0170550

20
The New York Times. (2020, January 28). Coronavirus World Map: Tracking the Global
Outbreak. Retrieved from https://www.nytimes.com/interactive/2020/world/coronavirus-
maps.html

Uddin, M., Mustafa, F., Rizvi, T. A., Loney, T., Suwaidi, H. A., Al-Marzouqi, A., Eldin, A. K.,
Alsabeeha, N., Adrian, T. E., Stefanini, C., Nowotny, N., Alsheikh-Ali, A., & Senok, A.
C. (2020). SARS-CoV-2/COVID-19: Viral Genomics, Epidemiology, Vaccines, and
Therapeutic Interventions. Viruses, 12(5), 526. https://doi-
org.libproxy1.usc.edu/10.3390/v12050526
Understanding America Study, uasdata.usc.edu/index.php.

21
TABLES
Table 1. Likelihood of Getting COVID-19 Vaccine (1=very unlikely, 4=very likely) Month
Comparison (June and October)

Month Mean Std Dev P-value
June 3.32 1.17 0.0572
October 3.23 1.22

Table 2. Vaccine Intent among L.A. County Residents, Split by Demographic
Demographic
% Will Get Vaccine
When Available
95% Confidence
Interval
P-Value
All 68.5 [65.7, 71.3]
Gender
Male 75.9 [71.7, 80.0] <0.0001
Female 63.6 [59.8, 67.4]
Age
18-30 years 60.7 [54.0, 67.4] <0.0001
31-40 years 64.8 [58.7, 71.0]
41-50 years 63.8 [56.9, 70.8]
51-64 years 76.9 [71.5, 82.3]
65+ years 76.9 [70.3, 83.5]
Household Income
$0-29,999 59.7 [53.9, 65.6] <0.0001
$30-59,999 62.3 [56.3, 68.2]
$60-99,999 70.9 [65.0, 76.9]
$100,000+ 81.1 [76.4, 85.8]
Race/Ethnicity
non-Hispanic White 77.7 [73.5, 82.0] <0.001
Hispanic/Latino 60.8 [56.1, 65.6]
Black/African
American 47.8 [37.4, 58.2]
Asian 73.8 [67.4, 80.2]

22
FIGURES

Figure 1. Likelihood of Getting Vaccine Distribution. This graph shows how participant
responses were distributed for the question “How likely are you to get a vaccine once one
becomes available?” (1= very unlikely, 4= very likely).

23

Figure 2. COVID-19 Vaccine Intent vs Age. This graph shows the Lowess smoothing curve of
the logistic regression model checking the linearity of intention of getting the COVID-19 vaccine
against the variable of age. Logistic regression analysis shows a positive correlation exists
between age and vaccine intent.

24

Figure 3. COVID-19 Vaccine Intent vs. Household Income. This graph shows the Lowess
smoothing curve of the logistic regression model checking the linearity of intention of getting the
COVID-19 vaccine against the variable of income level. Logistic regression analysis shows a
positive correlation exists between household income level and vaccine intent.

$12,500-
$14,999
<$5000 $35,000-
$49,999
$150,000<

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Sentiment analysis in the COVID-19 vaccine willingness among staff in the University of Southern California

PDF

Using social networks to recruit an HIV vaccine preparedness cohort

PDF

Characteristics associated with emergency medical services for cardiac arrest pre- and during COVID-19 in Los Angeles, CA

PDF

Examining the built and online environments to understand cannabis and e-cigarette availability, marketing, and product use in Southern California

PDF

A genome wide association study of multiple sclerosis (MS) in Hispanics

PDF

Factors influencing the decision and timing to undergo breast reconstruction after mastectomy for breast cancer in public hospital vs. private medical center from 2007 to 2013: a retrospective co...

PDF

The impact of the COVID-19 pandemic on cancer care delivery

PDF

Relationship between progression of atherosclerosis and coagulation measures in a randomized-controlled trial

PDF

Racial/ethnic differences in patient experiences with health care in association with the Gleason’s score level at prostate cancer diagnosis: findings from the SEER-CAHPS data

PDF

Association of comorbidity with prostate cancer tumor characteristics in African American men

PDF

Mixed-method analysis for school-aged children with access and functional needs during the Covid-19 pandemic

PDF

Associations between inflammatory markers and change in cognitive endpoints

PDF

Association between endogenous sex hormone levels and cognition: the Women’s Isoflavone Soy Health (WISH) trial

PDF

Anxiety symptoms and nicotine use among adolescents and young adults

PDF

The association of asthma and asthma-related medications on subclinical atherosclerosis: a cross-sectional analysis of four randomized clinical trials

PDF

The interplay between tobacco exposure and polygenic risk score for growth on birthweight and childhood acute lymphoblastic leukemia

PDF

Fine-grained analysis of temporal and spatial differences of behavior patterns and their correlation with the spread of COVID-19 in Los Angeles County

PDF

Native American ancestry among Hispanic Whites is associated with higher risk of childhood obesity: a longitudinal analysis of Children’s Health Study data

PDF

Relationship of blood pressure and antihypertensive medications to cognitive change in the BVAIT, WISH, and ELITE clinical trials

PDF

The effect of renal function on toxicity of E7389 (eribulin) among patients with bladder cancer

Asset Metadata

Creator Karpowicz, Nicole (author)

Core Title Relationship between L.A. County residents' demographics and willingness to take the COVID-19 vaccine

School Keck School of Medicine

Degree Master of Science

Degree Program Biostatistics

Publication Date 11/15/2021

Defense Date 11/14/2021

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

Tag COVID-19,demographics,OAI-PMH Harvest,pandemic,preventative measures,vaccine

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor De La Haye, Kayla (committee chair), Baezconde-Garbanati, Lourdes (committee member), Mack, Wendy (committee member)

Creator Email karpowic@usc.edu,nkarpowicz162@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC17138481

Unique identifier UC17138481

Legacy Identifier etd-KarpowiczN-10220

Document Type Thesis

Rights Karpowicz, Nicole

Type texts

Source 20211115-wayne-usctheses-batch-896-nissen (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

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

Repository Name University of Southern California Digital Library

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

Repository Email cisadmin@lib.usc.edu