University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

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

by
Yuwei Wu

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
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)

May 2023

Copyright 2023 Yuwei Wu

ii
Acknowledgments
I would like to express my heartfelt gratitude to my thesis committee chair, Dr. Mariana C. Stern,
for her careful guidance, professional knowledge, and invaluable support throughout the entire
process of completing my master's thesis.
I would also like to thank my thesis committee members, Dr. Albert Farias and Dr. Jennifer Tsui,
for their insightful feedback and valuable suggestions.
Finally, I would like to express my gratitude to my family for providing me with a nurturing
environment filled with happiness and love, which allowed me to grow and develop into the person
I am today.

iii
Table of Contents
Acknowledgments........................................................................................................................... ii
List of Tables .................................................................................................................................. iv
Abstract ........................................................................................................................................... v
Chapter 1 Introduction .................................................................................................................... 1
Chapter 2 Methods .......................................................................................................................... 5
2.1 Study population ....................................................................................................... 5
2.2 Variables .................................................................................................................... 5
2.3 Outcomes .................................................................................................................. 6
2.4 Statistical analysis ..................................................................................................... 7
Chapter 3 Results ............................................................................................................................ 9
3.1 Demographic and key clinical characteristics of patients ......................................... 9
3.2 Determinants of time between diagnosis and first treatment .................................. 10
3.3 Determinants of type of first intervention among breast cancer patients ............... 13
Chapter 4 Discussion .................................................................................................................... 15
References ..................................................................................................................................... 21

iv
List of Tables
Table 1. Variables abstracted from charts ..................................................................................... 25
Table 2. Demographic and key clinical characteristics of patients ............................................... 27
Table 3. Univariate determinants of time between diagnosis and treatment ................................ 31
Table 4. Multivariate determinants of time between diagnosis and first treatment ...................... 32
Table 5. Determinants of first type of treatment for breast cancer patients .................................. 33

v
Abstract
Purpose
The purpose of the study is to investigate the impact of the COVID-19 pandemic on cancer care
delivery to identify the most critical changes that may have occurred due to the pandemic, which
may have had the largest impact on patient outcomes. We compared the differences in cancer care
delivery for patients who started cancer treatment during the COVID-19 pandemic in 2020 with
those who started treatment in the same time period in 2019, focusing on identifying key patient-
level determinants of specific treatment patterns.
Methods
Using the NCCC and LAC clinical data warehouses, we conducted a chart review of patients who
started their treatment for breast, lung, or colorectal cancer at the Norris Comprehensive Cancer
Center-Keck Medical Center (NCCC) or the Los Angeles County Medical Center (LAC) from
March – September 2019 and those who started treatment in the same period in 2020. A total of
355 patients were included in the analysis. Patient demographic and clinical characteristics,
including treatment information, were abstracted from charts. Descriptive statistics was used to
examine the distributions of demographics and clinical characteristics of patients, comparing 2019
to 2020 patients. Univariate and multivariate logistic regression were used to examine the potential
determinants of time between diagnosis and the date of the first treatment for all cancers and for
each type of cancer, considering surgery, drug therapies, and radiation therapies. Polytomous
logistic regression was used to identify potential determinants of the type of first intervention for
breast cancer patients.
Results
The main difference in patient characteristics was a greater proportion of patients with stage at

vi
diagnosis > 3 in 2020 versus 2019, especially among breast cancer patients. When comparing
patients who started treatment in 2020 to those in 2019, we observed a higher average number of
telemedicine visits (3.8 vs 0.9), a shorter time between diagnosis and date of the first treatment
(35.5 days vs 53.3 days), a higher proportion of neoadjuvant therapy (29.3% vs 18.5%), and
reduced number of imaging scans (6.9 vs 8.8). In multivariate analyses of time between diagnosis
and first treatment, we observed that facility was a key determinant of delays (LAC vs NCCC OR
= 2.55; 95% CI: 1.37-4.75, p = 0.003), treatment year and stage and comorbidity were determinants
of shorter lag between diagnosis and treatment (2020 vs 2019 OR = 0.47; 95% CI: 0.29-0.77, p =
0.002, stage > 3 vs stage 0-2 OR = 0.39; 95% CI: 0.22-0.69, p = 0.001. Among breast cancer
patients, we observed reduced utilization of surgery as first line of treatment in 2020 (59%)
compared to 2019 (74.9%), with stage greater than 3 being the key determinant.
Conclusions
Our findings highlight key changes in cancer care delivery and factors associated with care
outcomes during the COVID-19 pandemic at the USC Norris Comprehensive Cancer Center and
the Los Angeles County Hospital. As expected, the number of telemedicine visits increased
significantly in 2020, and surprisingly, the time between diagnosis and the date of the first
treatment was shorter in 2020 compared to 2019, likely driven by prioritization of diagnosis and
treatment of patients with more advanced cancer. Comparing 2020 to 2019 key observed
differences included higher proportion of patients receiving neoadjuvant therapy, reduced imaging
scans, and reduced utilization of surgery among breast cancer patients as first line of treatment,
with the latter being driven by higher proportion of advanced stage at diagnosis. Overall, the
COVID-19 pandemic has had a significant impact on cancer care, and it will be important to
continue monitoring its impact on cancer outcomes because of these changes in treatment

vii
modalities.

1
Chapter 1 Introduction
On March 11th, 2020, the World Health Organization declared the emergence of the
coronavirus disease 2019 (COVID-19) as a pandemic of global proportions
[1]
. The impact of the
pandemic has been far-reaching and has brought forth a range of unprecedented challenges and
subsequent impacts, affecting various industries and sectors, including healthcare systems.
According to the World Health Organization (WHO), as of January 2023, the number of confirmed
COVID-19 cases globally has surpassed 750 million, with over 6.8 million reported fatalities
[2]
. In
response to the COVID-19 outbreak, many countries took measures to curb its spread, including
the closure or suspension of non-essential medical services. The impact of the pandemic on the
utilization of healthcare services has since become a topic of concern, with people worried about
the influence it may have had on access to essential medical care
[3]
.
Cancer care is a crucial aspect of healthcare as cancer is a disease that has a significant impact
on human health and is one of the leading causes of death worldwide. The effects of cancer on
individuals, families, and society cannot be denied. Early detection, diagnosis, and treatment of
cancer can significantly improve the chances of survival and the quality of life of those diagnosed.
However, the COVID-19 pandemic has had an impact on the diagnosis and treatment of cancer
[4]
,
such as imposing delays or cancellations of cancer screening and diagnosis for many patients
[5]
. A
recent narrative review of seven countries by the Global Consortium for Modeling COVID-19 and
Cancer highlighted how COVID-19-related disruptions may affect breast cancer screening
[6]
. With
health systems under strain, delays and disruptions caused by the pandemic directly or indirectly

2
impacted screening and diagnosis of cancer patients
[7]
. This is due to challenges imposed by the
pandemic on medical resources and staffing, given that many of these resources and manpower
had to be allocated to deal with the COVID-19 pandemic, with reduction of medical resources and
personnel for cancer care, thus reducing the chances of patients receiving timely diagnosis
[8]
.
Moreover, patients were worried about the risk of contracting COVID-19 when going to the
hospital, so some chose to delay or cancel regular cancer screening or use telemedicine instead.
In addition to delays in diagnosis, another important challenge faced by cancer patients was
disparities and delays in treatment. Access to cancer care continues to be a significant challenge in
many parts of the world, with disparities in healthcare infrastructure and resources leading to
unequal access
[9]
. These disparities became even more pronounced during the COVID-19
pandemic, when in order to prioritize medical resources to deal with the epidemic, many countries
and regions suspended or reduced the scope of medical services, with medical resources for cancer
treatment being inevitably affected
[10, 11, 12, 13]
. Cancer patients were also at higher risk for poor
COVID-19 outcomes, and many did not adhere to treatment or delayed treatment due to fears of
getting infected with COVID-19. In light of this situation, it is imperative that we understand the
impact of the COVID-19 pandemic on cancer treatment and outcomes.
By 2025, it is estimated that there will be an annual increase of more than 20 million new
cancer cases
[14]
. Among these cancer cases, breast and lung cancer are the most diagnosed cancers,
with lung cancer remaining the leading cause of cancer morbidity and mortality worldwide
[15]
.
Colorectal cancer, as the third most common malignancy and the second most deadly cancer, also

3
has a significant impact on human health
[16]
. These three cancers are also some of the most
common cancers in Los Angeles County, and their burden on the population is significant. Breast
cancer is the most diagnosed cancer in women in Los Angeles County, with an incidence rate of
127.6 cases per 100,000 women per year
[17]
. Lung cancer is the leading cause of cancer death in
both men and women in Los Angeles County. In 2020, it was estimated that there would be 5,040
new cases of lung cancer in the county, and 3,570 deaths from the disease
[17]
. Colorectal cancer is
the third most commonly diagnosed cancer in both men and women in Los Angeles County, with
an incidence rate of 39.1 cases per 100,000 people per year
[17, 18]
. These cancers have a significant
impact on the individuals in Los Angeles County, as well as on the healthcare system. In this study,
we focused on the impact of COVID-19 on cancer care delivery for these three of the top cancers
that affect our patient population.
Studying the impact of the COVID-19 pandemic on cancer characteristics and treatment
patterns may allow to identify the most critical changes in cancer care delivery that had the largest
impact in patient outcomes, while identifying those that did not. This knowledge may be valuable
in guiding changes in the future in response to new waves of COVID-19 or other pandemics, and
also informing next steps to catch-up treatment lost during COVID-19, and to address inequities
among patient groups more likely to not have received optimal care
[19]
. In this study, we report on
key differences in cancer care delivery for patients who started treatment at the USC Norris
Comprehensive Cancer Center (NCCC) or Los Angeles County + USC Medical Center (LAC)
between the start of the COVID-19 pandemic in March 2020 until September 2020 and compare

4
them to patients who started their treatment in the same time in 2019. We identified and report on
key differences in care delivery between these two periods and identify key determinants of
specific treatment patterns. We further discuss potential solutions to mitigate the impact of future
pandemics on cancer care by analyzing the challenges faced by patients and healthcare systems
during the COVID-19 pandemic.

5
Chapter 2 Methods
2.1 Study population
1.1 Using the NCCC and LAC clinical data warehouses, we identified eligible patients by
obtaining all records of patients with breast cancer, lung, or colorectal cancer codes entered into
the EMR for receiving any type of cancer care at NCCC or LAC between March – September
2019 or the same period in 2020. We evaluated all these cases to identify those that met the
following inclusion criteria: being 18 years of age or older and having started their treatment within
March-September 2020 or within March-September 2019. Exclusion criteria were patients
younger than 18 years of age and patients whose EMR entry indicated follow-up care, diagnosis,
or any other reason other than start of cancer treatment. For each eligible patient, the date of
diagnosis was recorded. We performed medical chart review to identify those patients who met the
eligibility criteria and conducted medical chart abstraction among those. A total of 355 patients
met the inclusion criteria and were included in the chart abstraction and subsequent analysis.
2.2 Variables
Variables were abstracted from medical charts (Table 1). Patient characteristics included date
of birth, sex, race, and ethnicity. We considered the following clinical characteristics for all cancers:
treatment hospital, cancer type, stage at diagnosis, date of diagnosis, and date of the first treatment,
including surgical procedures, drug therapies or radiation therapy, total provider encounters within
the period of the study, total provider encounters within the period of the study, total provider
encounters in the first 90 days since the start of treatment, number of telemedicine visits, total

6
hospitalizations, total hospitalizations for cancer surgery, date of clinic visits, first intervention,
dose reductions in first 90 days, the reason for dose changes, receipt of neoadjuvant therapy (yes
or no), adjuvant therapy, date of start of chemotherapy, date of start of radiation therapy, last date
of radiation treatment (dose), date of last ED visit, date of imaging scan, infected with COVID-19
(yes/no), type of comorbidity, and comorbidity (yes/no). We also considered the following clinical
characteristics for breast cancer: type of surgery, use of nonadjunctive endocrine therapy, and
considered the change of chemotherapy protocol from IV to oral for lung cancer and colorectal
cancer.
We developed several variables related to time (in days) for all cancers. We measured the time
between diagnosis and the date of first treatment, the time between diagnosis date and the start
date of chemotherapy, and the time between diagnosis and the start of radiation therapy, for patients
who received these treatments. We also tracked the total number of imaging scans and the number
of visits within the first six months since diagnosis. In addition, for breast cancer, we measured the
time between surgery and the first post-surgery mammogram, the time between diagnosis and
reconstructive surgery date, and the time between surgery date and the actual reconstructive
surgery. For colorectal cancer, we measured the time between diagnosis and the date of radiation,
the time between surgery and the actual colonoscopy.
2.3 Outcomes
Our main outcome we inquired about was the time between diagnosis and the date of the first
treatment for each type of cancer and for all cancers combined. To capture this outcome, we

7
generated dichotomous variables that used the median time between diagnosis and date of the first
treatment among 2019 patients for each type of cancer as cancer-specific cut-points, with the
median for breast cancer being 52.5 days, for lung cancer 27 days, and for colorectal cancer 42
days. For all cancers combined, we also used the cancer-specific cut-points to generate a
dichotomous variable. Patients were labeled as shorter time if the time between diagnosis and
treatment were equal to or less than the median, and longer time if above the median. In all
instances, we compared patients with time between diagnosis and treatment equal or below the
median, to those above the median.
We also investigated as an outcome the type of first intervention for breast cancer patients.
We created a new variable with three categories: "Surgery," "Drug therapy," and "Other" (including
radiation, palliative care, or no therapy).
2.4 Statistical analysis
Descriptive statistics were used to examine the distributions of the demographics and clinical
prognostic and treatment characteristics of patients. We compared the variables described above
between 2019 and 2020 patients using frequencies and Chi-square tests for categorical variables
and t-tests for continuous variables to identify factors associated with statistical differences since
the pandemic.
To identify potential determinants of time between diagnosis and date of the first treatment
for all cancers combined and for each type of cancer, we used univariate and multivariate logistic
regression models, estimating Odds ratios (OR) and 95% confidence intervals (CI). We considered

8
the following potential determinants: treatment hospital, treatment year, age, sex, race and ethnicity,
cancer type, stage, type of comorbidity, and comorbidity (yes/no) for all cancers combined; and
treatment hospital, treatment year, age, sex, race and ethnicity, stage, type of comorbidity, and
comorbidity (yes/no) for each type of cancer.
To identify potential determinants of the type of first intervention for breast cancer patients,
we used polytomous logistic regression and estimated Odds ratios (OR) and 95% confidence
intervals (CI). We considered the following possible determinants comparing cases that received
surgery as the first treatment to those who received drug therapy, or other therapies (radiation,
palliative care, or no therapy): treatment hospital, age, sex, race and ethnicity, and stage.
In statistical analyses, p < 0.05 was considered statistically significant. All statistical analyses
were performed using STATA software (version 17.0).

9
Chapter 3 Results
3.1 Demographic and key clinical characteristics of patients
We identified a total of 2770 patients who had received any type of cancer care between
March and September 2020 or 2019 (1064 breast cancer, 1152 lung cancer, and 554 colorectal
cancer). Among these, we identified a total of 355 patients who had started their cancer care for
breast, lung, or colorectal cancer during these time periods (193 in 2019 and 162in 2020). Among
these, we identified a total of 189 breast cancer cases, 70 lung cancer cases, and 96 colorectal
cancer cases (Table 2). When comparing patients who started treatment between March-September
2020 to the same period in 2019, we observed several key differences of statistical significance
(Table 2). When considering all tumor types combined, we observed that there was a slightly higher
number of the stage at diagnosis >2 diagnoses in 2020 compared to 2019, and these differences
were or borderline statistical significance (p = 0.06). We also observed a higher average number
of telemedicine visits in 2020 [mean: 3.8, SD:(3.7)] versus 2019 [0.9 (1.9)], a shorter time between
diagnosis and date of the first treatment in 2020 [35.5 days (38.1)] versus 2019 [53.3 days (39.5)],
a greater proportion of utilization of neoadjuvant therapy in 2020 (29.3%) compared to 2019
(18.5%), and less number of imaging scan times in 2020 [6.9 (5.3)] compared to 2019 [8.8 (8.3)]
(Table 2).
When considering each cancer type separately, we observed similar patterns between 2019
and 2020 patients, with some salient differences. Among breast cancer patients, in addition to a
greater number of telemedicine visits in 2020 [4.4 (4.3)] compared to 2019 [0.8 (1.2)], we also

10
observed that there was a greater proportion of more advanced stages at diagnosis during 2020
compared to 2019 (15.2% stage IV in 2020 versus 6.5% in 2019, p = 0.023), a higher number of
total provider encounters in 2020 [13.6 (9.7)] compared to 2019 [10.3 (10.1)] (p = 0.027). We also
observed a shorter time between diagnosis and treatment in 2020 [42.1 days (36.5)] compared to
2019 [56.6 days (32.5)] (p = 0.005), which seemed driven by shorter times between diagnosis and
start of chemotherapy [67 days (45.7) in 2020 versus 91.2 days (56.3) in 2019, p = 0.021]. We also
observed there was a shorter time between surgery and date of reconstructive surgery in 2020 [15
days (78.5)] versus 2019 [130.1 days (180.6)] (p = 0.051) (Table 2).
Among lung cancer patients, we observed that there was a smaller proportion of patients
receiving treatment (14%) at LAC during 2020 compared to 2019 (38%), a higher number of total
hospitalizations for cancer surgery in 2020 (0.6 + 0.6) compared to 2019 [1.8 (1.7)], and a greater
proportion of utilization of adjuvant therapy (27%) in 2020 compared to 2019 (7%) (p = 0.031)
(Table 2). Among colorectal cancer patients, we also observed there were a shorter time between
diagnosis and date of the first treatment in 2020 [33.3 days (30.8)] versus 2019 [55.5 days (53.0)]
(p = 0.019), and a shorter time between diagnosis and date of start of radiation therapy in 2020
[114.7 days (75.8)] compared to 2019 [324.8 days (404)] (p = 0.03) (Table 2).
3.2 Determinants of time between diagnosis and first treatment
Given that we observed statistically significant differences in the time between diagnosis and
the date of first treatment, we evaluated potential determinants for all cancers combined. We
considered the following potential determinants: treatment hospital, treatment year, age, sex, race

11
and ethnicity, stage, cancer type, and comorbidity (yes/no).
Using univariate analysis, for all cancers combined, we observed that receiving treatment at
LAC, and being NHB and Hispanic were associated with greater chance of having longer time
between diagnosis and treatment (Table 3). Instead, those patients receiving treatment in 2020,
female patients, having advanced stage (stage >3) (vs. stage 0-2), and without comorbidity were
also associated with shorter time.
Based on the findings from univariate analyses, we evaluated possible determinants using
multivariate models, including treatment hospital, treatment year, sex, race and ethnicity, stage,
cancer type and comorbidity (yes/no) (Table 3). We observed that patients who started their
treatment at LAC were 2.55 times more likely to have longer time between diagnosis and treatment
than patients who received treatment at NCCC (95%CI: 1.37-4.75, p = 0.003). NHB and Hispanic
patients had higher likelihood of longer time between diagnosis and treatment than NHW patients;
however, these differences were not statistically significant. Instead, patients who received
treatment in 2020 were less likely to show longer time between diagnosis and first treatment
compared to those patients who received treatment in 2019 (OR = 0.47; 95% CI: 0.29-0.77, p =
0.002). Patients with stage >3 were less likely to show longer time between diagnosis and first
treatment compared to patients with stage 0-2 (OR = 0.39; 95% CI: 0.22-0.69, p = 0.001).
We also evaluated potential determinants of the time between diagnosis and the date of the
first treatment by cancer type, considering same possible determinants. When considering breast
cancer patients, we observed that receiving treatment at LAC compared to receiving treatment at

12
NCCC and being Hispanic were associated with greater chance of having longer time between
diagnosis and treatment (>52.5 days) in univariate analysis. Receiving treatment in 2020 compared
to 2019 and the advanced stage (stage >3) compared to stage 0-2 were associated with shorter time
(Table 3). In multivariate models, we observed that patients who started their treatment at LAC
were 4.18 times more likely to have longer time between diagnosis and treatment than patients
who received treatment at NCCC (95% CI: 2.14-8.17, p < 0.001) (Table 4). Patients who received
treatment in 2020 were instead less likely to show longer time between diagnosis and first
treatment compared to those patients who received treatment in 2019 (OR = 0.34; 95% CI: 0.17-
0.68, p = 0.002). Patients with stage >3 were less likely to show longer time between diagnosis
and first treatment compared to early stage (stage 0-2) patients (OR = 0.41; 95% CI: 0.17-0.95, p
= 0.038).
Among lung cancer patients, we observed that receiving treatment at LAC, being Hispanic
were associated with greater chance of having longer time between diagnosis and treatment (>27
days) in univariate analysis (Table 3). In multivariate models, we observe that receiving treatment
at LAC was 5.72 times (95% CI: 0.83-39.26, p = 0.076) more likely to have longer time between
diagnosis and treatment (> 27 days) than receiving treatment at NCCC, respectively. In this
multivariate model, being Hispanic was no longer seemed associated with time between diagnosis
and first treatment (Table 4).
Among colorectal cancer patients, we observed that increasing age was associated with
greater chance of having longer time between diagnosis and treatment (>42 days) in univariate

13
analysis (Table 3). Instead, female patients, patients with advanced stage (stage >3) and without
comorbidity were associated with shorter time. In multivariate models, we confirmed that female
patients were less likely to show longer time between diagnosis and first treatment compared to
male patients (OR = 0.39; 95% CI: 0.16-0.98, p = 0.046), patients with stage >3 were still less
likely to show longer time between diagnosis and first treatment compared to early-stage (stage 0-
2) patients (OR = 0.31; 95% CI: 0.12-0.81, p = 0.017), and patients without comorbidity were less
likely to show longer time between diagnosis and first treatment compared to patients with
comorbidity (OR = 0.26; 95% CI: 0.12-1.04, p = 0.06) (Table 4).
3.3 Determinants of type of first intervention among breast cancer patients
Given the finding of the greater proportion of drug therapy as the first line of treatment among
breast cancer patients who started treatment in 2020 (33.3%) compared to 2019 (21.5%), and the
reduced proportion of surgery in 2020 (59%) compared to 2019 (74.8%) as the first line of therapy,
we evaluated possible determinants of this treatment pattern. We considered the following possible
determinants comparing cases that received surgery as the first treatment to those who received
either drug therapy, or other therapies (radiation, palliative care, or no therapy): treatment hospital,
age, sex, race and ethnicity, and stage.
In multivariate analyses, when comparing receipt of drug therapy or receipt of other therapy
as the first intervention to receipt of surgery, we observed that patients with stage > 3 were 25.37
times more likely to receive drug therapy compared to patients with stage 0-2 (95% CI: 8.55-75.32,
p < 0.001), patients with stage > 3 were 214.92 times more likely to receive drug therapy compared

14
to stage 0-2 patients (95% CI: 19.59-2358.48, p < 0.001). In this multivariate model, even though
we observed that those receiving treatment at LAC had higher likelihood of receiving drug therapy
than those who received treatment at NCCC, and that NHB, Hispanic, Asian/PI, and other patients
had higher likelihood of receiving drug therapy than NHW patients, these differences were not
statistically significant. Similarly, even though we observed that patients receiving treatment at
LAC were less likely to receive other therapies than surgery or drug therapy than those who
received treatment at NCCC, and that NHB, Hispanic, and Asian /PI patients had higher likelihood
of receiving other therapy than NHW patients, these differences were also not statistically
significant (Table 6).

15
Chapter 4 Discussion
In this study, we compared the differences in cancer care delivery among patients who started
treatment during the COVID-19 pandemic in 2020 with those started treatment in the same time
period in 2019, at two large medical centers in Los Angeles, and we also identified key
determinants of specific treatment patterns.
For all cancer types combined, we found that number of telemedicine visits increased in 2020
compared to 2019, which is consistent with the increase in telemedicine oncology consultations
reported among all cancer types by TPMG (The Permanente Medical Group)
[20]
. Telemedicine
helped reduce the number of in-person visits to medical facilities and the risk of contracting
COVID-19
[21]
.
The proportion of neoadjuvant therapy was higher in 2020 compared to 2019, which is
consistent with prior studies in the Hospital of the University of Pennsylvania (HUP) and Mayo
Clinic
[22, 23]
. Neoadjuvant therapy refers to treatment given before the main therapy, such as surgery,
with the goal of shrinking or eradicating cancer before the primary treatment. The observed
increase in neoadjuvant therapy might be driven by the desire or need to reduce the number of
hospital visits for cancer patients during the pandemic
[24]
, and/or limitations in the use of operating
rooms at the height of the pandemic. Furthermore, there may be concerns about the risk of
contracting COVID-19 while in the hospital, which may lead to a preference for neoadjuvant
therapy instead of surgery or other primary treatments.
The number of imaging scan times was less in 2020 compared to 2019, which is also

16
consistent with prior studies that show that many patients, and/or physicians, opted to postpone
many imaging exams during COVID-19
[22, 25]
. The decreases in cancer screening during the early
phases of the COVID-19 pandemic still did not resolve after the initial pandemic surges
[26, 27]
.
While a decrease in imaging scans may have benefits such as reducing the risk of exposure to
COVID-19, it may also result in delays in cancer diagnosis and treatment, which may explain our
observation of slightly higher advanced stage at diagnosis in 2020 compared to 2019, especially
for breast cancer patients, among whom we observed a greater proportion of patients with more
advanced stages at diagnosis during 2020 compared to 2019.
The shorter time between diagnosis and the date of the first treatment in 2020 compared to
2019 was unexpected. Because of patients’ reluctance to come to the hospital for fear of the risk
of COVID infection, and the strain on medical resources during the COVID-19 pandemic
[21]
, we
expected the time between diagnosis and first treatment to be longer during the pandemic. When
considering determinants among all cancer types combined, we observed that the time between
diagnosis and treatment seemed affected by mainly the treatment facility and diagnosis stage.
Patients who received treatment at LAC were more likely to have a longer time between diagnosis
and first treatment than patients who received treatment at NCCC. This finding is consistent with
the fact that LAC is a safety-net hospital that receives a higher volume of patients and includes a
large emergency department that received many COVID-19 cases, and thus was severely impacted
by the pandemic. Instead, we observed that patients who received treatment in 2020 were less
likely to show longer time between diagnosis and first treatment compared to those patients who

17
received treatment in 2019, and patients with stage >3 were less likely to show a longer time
between diagnosis and first treatment compared to early-stage patients, which likely reflects the
prioritization of higher risk patients at the time of scheduling, given the limitations of patient
volume during the pandemic. We also speculate that the increased proportion of people receiving
neoadjuvant therapy contributed to shorter time between diagnosis and first treatment. The
increased number of telemedicine visits in 2020 also may also have contributed to shorter time
between diagnosis and first treatment, as patients had more flexibility to see their providers. Lastly,
as mentioned above, likely due to cancer diagnostic delays, there was a proportion of patients being
diagnosed at more advanced stages in 2020 compared to 2019, who require more timely treatment.
As well as delays in diagnostic procedures for patients with no symptoms, which may have resulted
in fewer patients with early-stage being diagnosed. These facts may also have contributed to the
shorter time between diagnosis and treatment.
Our observation that breast cancer patients with stage >3 were more likely to receive drug
therapy or other therapy, rather than surgery, is consistent with a previous report showing that the
choice of initial therapy for newly diagnosed patients shifted from surgery toward hormonal
therapy during the COVID-19 pandemic
[23]
. This is likely because advanced breast cancer may
have spread beyond the breast and lymph nodes to other parts of the body, making it more difficult
to treat with surgery alone
[28]
. In such cases, drug therapy, palliative care, radiation, or no therapy
may be recommended to manage symptoms, relieve pain, and improve the quality of life for
patients with advanced breast cancer
[29]
.

18
Our study has several strengths. First, consideration of three top cancers that affect our patient
population in Los Angeles County, utilizing comprehensive data from two medical facilities that
serve two different populations: an insured population of middle to high-income, and a safety-net
population, albeit both treated by the same medical professionals. Second, utilization of medical
chart abstraction considering several key clinical patterns. Our study also has several limitations.
First, because our study only includes patients from NCCC and LAC, the results may not be
generalizable to other institutions or regions; however, given the range of patient characteristics
across these two institutions, they are likely to be reasonable representative of the Los Angeles
County population. Second, we did not examine yet the impact of the observed treatment patterns
on cancer outcomes, such as cancer recurrence rates or survival rates, which will provide a more
comprehensive picture of the impact of the pandemic on cancer care. These analyses will follow-
up those presented in this study. Third, even though we considered as many factors as we could
base on the data that was available for abstraction, we cannot discard the possibility that there
might be additional determinants or confounders not considered in our analyses, for example,
patient reported barriers and/or social determinants of health. Fourth, the sample size may have
limited our ability to detect associations that may actually exist. Fifth, our study only includes
patients who received treatment, and does not capture patients who may have been diagnosed with
cancer but never pursed treatment without our window of analysis, or, sought treatment elsewhere
due to changes in insurance coverage, relocation, or personal preferences.
Despite these limitations, our study highlights several challenges faced by patients and

19
healthcare systems during the COVID-19 pandemic, which may be useful in developing strategies
to mitigate the impact of future pandemics on cancer care. The use of telemedicine may be one
solution to help reduce the risk of infection and ensure continuity of care for cancer patients and
may have contributed to shorter times between diagnosis and treatment. However, it is important
to note that telemedicine may not be suitable for all patients, particularly those with low digital
literacy or non-English speakers, and some patients may still require or benefit from in-person
visits. It's also worth noting that telemedicine can encompass various forms of remote
communication, including videoconferencing, phone calls, and messaging. The choice of
technology may depend on the specific needs and capabilities of the patient and healthcare provider.
At LAC they used phone calls only for a long time during the pandemic, in contrast to mostly
video appointments used at NCCC. These differences may have also contributed to differences in
care delivery. In some cases, in-person visits may be necessary or preferred, while in other cases,
telemedicine may offer a convenient and effective alternative. Therefore, a combination of
telemedicine and in-person visits may be necessary to provide optimal cancer care during
pandemics. Despite the increased use of telemedicine, there is still a worrisome increase in higher
stage diagnosis, particularly among breast cancer patients. Therefore, further improvements are
needed to avoid delays in screening and diagnosis during a pandemic.
In conclusion, the COVID-19 pandemic has had a significant impact on cancer care, and it
will be important to continue monitoring its impact on cancer outcomes. Future research should
focus on identifying effective strategies to mitigate the impact of pandemics on cancer care,

20
building upon the successes and failures of the use of telemedicine and other innovative
approaches to ensure timely diagnoses. Moreover, further research is needed to understand the
longer-term impact of changes in care delivery during the COVID-19 pandemic, and strategies to
mitigate this impact. Ultimately, it is critical to ensure that cancer patients continue to receive
timely and appropriate care, even during times of crisis.

21
References
[1] Li, T., Nickel, B., Ngo, P ., McFadden, K., Brennan, M., Marinovich, M. L., & Houssami,
N. (2023). A systematic review of the impact of the COVID-19 pandemic on breast cancer
screening and diagnosis. The Breast, 67, 78–88. https://doi.org/10.1016/j.breast.2023.01.001
[2] WHO Coronavirus (COVID-19) Dashboard. (n.d.). WHO Coronavirus (COVID-19)
Dashboard | WHO Coronavirus (COVID-19) Dashboard With Vaccination Data.
https://covid19.who.int
[3] Moynihan, R., Sanders, S., Michaleff, Z. A., Scott, A. M., Clark, J., To, E. J., Jones, M.,
Kitchener, E., Fox, M., Johansson, M., Lang, E., Duggan, A., Scott, I., & Albarqouni, L. (2021,
March). Impact of COVID-19 pandemic on utilization of healthcare services: a systematic
review. BMJ Open, 11(3), e045343. https://doi.org/10.1136/bmjopen-2020-045343
[4] Patt, D., Gordan, L., Diaz, M., Okon, T., Grady, L., Harmison, M., Markward, N., Sullivan,
M., Peng, J., & Zhou, A. (2020, November). Impact of COVID-19 on Cancer Care: How the
Pandemic Is Delaying Cancer Diagnosis and Treatment for American Seniors. JCO Clinical
Cancer Informatics, 4, 1059–1071. https://doi.org/10.1200/cci.20.00134
[5] Wolfs, T. F., Attarbaschi, A., Balduzzi, A., Bernardo, M. E., Bomken, S., Borkhardt, A.,
Bourquin, J. P., Dufour, C., Gennery, A., Grainger, J., Hasle, H., Hrusak, O., Izraeli, S.,
Mechinaud, F., Trka, J., & Vormoor, J. (2020, September 11). COVID-19 – Impact on
Childhood Haematology Patients. HemaSphere, 4(5), e465.
https://doi.org/10.1097/hs9.0000000000000465
[6] Figueroa, J. D., Gray, E., Pashayan, N., Deandrea, S., Karch, A., Vale, D. B., Elder, K.,
Procopio, P., van Ravesteyn, N. T., Mutabi, M., Canfell, K., & Nickson, C. (2021, October).
The impact of the Covid-19 pandemic on breast cancer early detection and screening.
Preventive Medicine, 151, 106585. https://doi.org/10.1016/j.ypmed.2021.106585
[7] Li, T., Nickel, B., Ngo, P., McFadden, K., Brennan, M., Marinovich, M. L., & Houssami, N.
(2023, February). A systematic review of the impact of the COVID-19 pandemic on breast
cancer screening and diagnosis. The Breast, 67, 78–88.
https://doi.org/10.1016/j.breast.2023.01.001
[8] Cancer Screening & COVID-19. (n.d.). Cancer Screening & COVID-19.
https://www.cancer.org/healthy/find-cancer-early/cancer-screening-during-covid-19-
pandemic.html

22
[9] The State of Cancer Health Disparities in 2022 - CDPR22. (n.d.). Cancer Progress Report.
https://cancerprogressreport.aacr.org/disparities/cdpr22-contents/cdpr22-the-state-of-cancer-
health-disparities-in-2022/
[10] Cancer Facts & Figures 2021 – Special Section: COVID-19 and Cancer.
https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-
cancer-facts-and-figures/2021/special-section-covid19-and-cancer-2021.pdf
[11] Abdelkader, H., & El Kassas, M. (2022, May 19). Tailored treatment strategies for cancer
patients during COVID-19 pandemic. Reports of Practical Oncology and Radiotherapy, 27(2),
318–330. https://doi.org/10.5603/rpor.a2022.0024
[12] Bakouny, Z., Hawley, J. E., Choueiri, T. K., Peters, S., Rini, B. I., Warner, J. L., & Painter, C.
A. (2020, November). COVID-19 and Cancer: Current Challenges and Perspectives. Cancer
Cell, 38(5), 629–646. https://doi.org/10.1016/j.ccell.2020.09.018
[13] Llanos, A. A. M., Ashrafi, A., Ghosh, N., Tsui, J., Lin, Y ., Fong, A. J., Ganesan, S., & Heckman,
C. J. (2023, January 13). Evaluation of Inequities in Cancer Treatment Delay or
Discontinuation Following SARS-CoV-2 Infection. JAMA Network Open, 6(1), e2251165.
https://doi.org/10.1001/jamanetworkopen.2022.51165
[14] Zugazagoitia, J., Guedes, C., Ponce, S., Ferrer, I., Molina-Pinelo, S., & Paz-Ares, L. (2016,
July). Current Challenges in Cancer Treatment. Clinical Therapeutics, 38(7), 1551–1566.
https://doi.org/10.1016/j.clinthera.2016.03.026
[15] Tao, Z., Shi, A., Lu, C., Song, T., Zhang, Z., & Zhao, J. (2014, December 28). Breast Cancer:
Epidemiology and Etiology. Cell Biochemistry and Biophysics, 72(2), 333–338.
https://doi.org/10.1007/s12013-014-0459-6
[16] Xi, Y., & Xu, P. (2021, October). Global colorectal cancer burden in 2020 and projections to
2040. Translational Oncology, 14(10), 101174. https://doi.org/10.1016/j.tranon.2021.101174
[17] California Cancer Registry. (2020). Cancer incidence and mortality in California, 2020.
https://www.ccrcal.org/.
[18] American Cancer Society. (2020). Cancer facts and figures 2020.
https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-
figures-2020.html.

23
[19] Joung, R. H., Nelson, H., Mullett, T. W., Kurtzman, S. H., Shafir, S., Harris, J. B., Yao, K. A.,
Brajcich, B. C., Bilimoria, K. Y., & Cance, W. G. (2022, March 21). A national quality
improvement study identifying and addressing cancer screening deficits due to the COVID‐ 19 pandemic. Cancer, 128(11), 2119–2125. https://doi.org/10.1002/cncr.34157
[20] Liu, R., Sundaresan, T., Reed, M. E., Trosman, J. R., Weldon, C. B., & Kolevska, T. (2020,
June). Telehealth in Oncology During the COVID-19 Outbreak: Bringing the House Call
Back Virtually. JCO Oncology Practice, 16(6), 289–293. https://doi.org/10.1200/op.20.00199
[21] VANNI, G., MATERAZZO, M., PELLICCIARO, M., INGALLINELLA, S., RHO, M.,
SANTORI, F., COTESTA, M., CASPI, J., MAKAROVA, A., PISTOLESE, C. A., &
BUONOMO, O. C. (2020). Breast Cancer and COVID-19: The Effect of Fear on Patients’
Decision-making Process. In Vivo, 34(3 suppl), 1651–1659.
https://doi.org/10.21873/invivo.11957
[22] Tonneson, J. E., Hoskin, T. L., Day, C. N., Durgan, D. M., Dilaveri, C. A., & Boughey, J. C.
(2021, November 23). Impact of the COVID-19 Pandemic on Breast Cancer Stage at
Diagnosis, Presentation, and Patient Management. Annals of Surgical Oncology, 29(4), 2231–
2239. https://doi.org/10.1245/s10434-021-11088-6
[23] Hawrot, K., Shulman, L. N., Bleiweiss, I. J., Wilkie, E. J., Frosch, Z. A. K., Jankowitz, R. C.,
& Laughlin, A. I. (2021, September). Time to Treatment Initiation for Breast Cancer During
the 2020 COVID-19 Pandemic. JCO Oncology Practice, 17(9), 534–540.
https://doi.org/10.1200/op.20.00807
[24] Schrag, D., Hershman, D. L., & Basch, E. (2020, May 26). Oncology Practice During the
COVID-19 Pandemic. JAMA, 323(20), 2005. https://doi.org/10.1001/jama.2020.6236
[25] Boettcher, A. N., Hammoud, D. A., Weinberg, J. B., Agarwal, P., Mendiratta-Lala, M., &
Luker, G. D. (2020, November 1). Cancer Imaging and Patient Care during the COVID-19
Pandemic. Radiology: Imaging Cancer, 2(6), e200058.
https://doi.org/10.1148/rycan.2020200058
[26] Doan, C., Li, S., & Goodwin, J. S. (2023, February 3). Breast and Lung Cancer Screening
Among Medicare Enrollees During the COVID-19 Pandemic. JAMA Network Open, 6(2),
e2255589. https://doi.org/10.1001/jamanetworkopen.2022.55589
[27] Cheng, D., Ghoshal, S., Zattra, O., Flash, M., Lang, M., Liu, R., Lev, M. H., Hirsch, J. A.,
Saini, S., Gee, M. S., & Succi, M. D. (2023, February 12). Trends in oncological imaging

24
during the COVID ‐19 pandemic through the vaccination era. Cancer Medicine.
https://doi.org/10.1002/cam4.5678
[28] Surgery for Breast Cancer | Breast Cancer Treatment. (n.d.). Surgery for Breast Cancer | Breast
Cancer Treatment. https://www.cancer.org/cancer/breast-cancer/treatment/surgery-for-breast-
cancer.html
[29] Cherny, N. I., Paluch-Shimon, S., & Berner-Wygoda, Y. (2018, December). Palliative care:
needs of advanced breast cancer patients. Breast Cancer: Targets and Therapy, Volume 10,
231–243. https://doi.org/10.2147/bctt.s160462

25
Table 1. Variables abstracted from charts
Variables Description
All cancers
Year Year of treatment (2019 or 2020)
Site Treating hospital (NCCC or LAC)
Date of birth
Sex Male or female
Race
American Indian or Alaska Native, Asian, Black or
African American, Native Hawaiian or Other Pacific
Islander, White, Other
Ethnicity Hispanic or Non-Hispanic
Type of Cancer Breast cancer, lung cancer, colorectal cancer
Stage at diagnosis Stages 0-4 (AJCC staging)
Date of diagnosis
Dates of first treatment
Total provider encounters within the period of the
study
Total provider encounters in first 90 days since start
of treatment
Number of telemedicine visits
Total hospitalizations (any time)
Total hospitalizations for cancer surgery (any time)
Date of clinic visits
First intervention
Surgery, radiation, drug therapy, palliative care, no
therapy
Dose reductions in first 90 days
Reason for dose changes Toxicity, skipped visits, other
Total number of administered cycles in first 90 days
of IV therapy
Neoadjuvant therapy Yes/No
Adjuvant therapy Yes/No
Date of start of Chemotherapy
Date of start of radiation therapy
Last date of radiation treatment
Date of last ED visit
Date of imaging scan Date of imaging scan (CT, PET, MRI, Etc.)
Date of surgery
Has patient been infected with COVID-19? Yes/No
Type of Comorbidity
Hypertension, hyperlipidemia, osteoarthritis,
hypothyroidism, diabetes mellitus, coronary artery
disease, other
Comorbidity Yes/No
Breast cancer
Date of post-surgery mammogram
Date of first chemotherapy
Type surgery Mastectomy, lumpectomy
Date of reconstructive surgeries and prophylactic
mastectomies
Use of neoadjuvant endocrine therapy Yes/No
Colorectal cancer
Date of actual colonoscopy Date of actual colonoscopy

26
Date of radiation Date of radiation for colorectal cancer
Change of chemotherapy protocol from IV to oral? Yes/No
Lung cancer
Date of radiation Date of radiation for lung cancer
Change of chemotherapy protocol from IV to oral? Yes/No

27

Table 2. Demographic and key clinical characteristics of patients
All cancers combined Breast cancer Lung cancer Colorectal cancer
Variables* 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value
n 193 162 109 80 34 36 50 46
Site 0.61 0.23 0.02 0.52
NCCC
114
(59.1%)
100
(61.7%)

68
(62.4%)
43
(53.8%)

21
(62%)
31
(86%)

25
(50%)
26
(57%)
LAC
79
(40.9%)
62
(38.3%)

41
(37.6%)
37
(46.3%)

13
(38%)
5 (14%)
25
(50%)
20
(43%)
Age
64.0
(11.9)
62.5
(13.6)
0.27
62.5
(10.7)
59.7
(12.0)
0.098
69.0
(12.1)
70.2
(11.9)
0.70
64.2
(13.4)
61.6
(15.3) 0.37
Sex 0.27 0.39 0.71 0.52
Male
45
(23.4%)
46
(28.6%)
1 (0.9%) 2 (2.5%)
19
(56%)
18
(51%)

25
(50%)
26
(57%)
Female
147
(76.6%)
115
(71.4%)

107
(99.1%)
78
(97.5%)

15
(44%)
17
(49%)

25
(50%)
20
(43%)
Race/ethnicity 0.93 0.85 0.47 0.85
NHW
46
(25.3%)
41
(26.6%)

30
(28.6%)
20
(25.3%)
4 (13%) 9 (28%)
12
(26%)
12
(28%)
NHB 9 (4.9%) 8 (5.2%) 2 (1.9%) 2 (2.5%) 4 (13%) 2 (6%) 3 (6%) 4 (9%)
Hispanic
68
(37.4%)
62
(40.3%)

41
(39.0%)
36
(45.6%)
9 (30%) 9 (28%)
18
(38%)
17
(40%)
Asian/PI
53
(29.1%)
39
(25.3%)

28
(26.7%)
17
(21.5%)

12
(40%)
12
(38%)

13
(28%)
10
(23%)
Other 6 (3.3%) 4 (2.6%) 4 (3.8%) 4 (5.1%) 1 (3%) 0 (0%) 1 (2%) 0 (0%)
Type of cancer 0.37
Breast
Cancer
109
(56.5%)
80
(49.4%)

Lung
Cancer
34
(17.6%)
36
(22.2%)

Colorectal
Cancer
50
(25.9%)
46
(28.4%)

Stage at diagnosis 0.06 0.023 0.36 0.40
Stage 0-1
77
(40.1%)
43
(26.7%)

65
(60.2%)
32
(40.5%)
3 (9%) 8 (22%) 9 (18%) 3(7%)
Stage 2
37
(19.3%)
38
(23.6%)

27
(25.0%)
22
(27.8%)
3 (9%) 3 (8%) 7 (14%)
13
(28%)

Stage 3
35
(18.2%)
40
(24.8%)
9 (8.3%)
13
(16.5%)

10
(29%)
6 (17%)
16
(32%)
21
(46%)

Stage 4
43
(22.4%)
40
(24.8%)
7 (6.5%)
12
(15.2%)

18
(53%)
19
(53%)

18
(36%)
9 (20%)

28

All cancers combined Breast cancer Lung cancer Colorectal cancer
Variables* 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value
Total provider encounters within the period of the study
8.7 (8.4) 10.2 (8.2) 0.099
10.3
(10.1)
13.6 (9.7) 0.027 5.6 (3.2) 7.5 (4.7) 0.059 7.4 (5.8) 6.1 (3.5) 0.23
Total provider encounters in first 90 days since start of treatment
5.1 (4.4) 5.6 (3.6) 0.32 5.7 (4.5) 6.8 (4.3) 0.085 3.9 (2.1) 4.6 (2.2) 0.16 4.8 (4.9) 4.0 (2.0) 0.37
Number of telemedicine visits
0.9 (1.9) 3.8 (3.7) <0.001 0.8 (1.2) 4.4 (4.3) <0.001 2.1 (3.6) 3.6 (3.4) 0.094 0.5 (1.1) 3.1 (2.6) <0.001
Total hospitalizations (any time)
1.3 (1.6) 1.1 (1.1) 0.42 0.7 (1.0) 0.9 (1.1) 0.41 1.8 (1.7) 1.2 (1.1) 0.11 2.1 (2.0) 1.6 (0.9) 0.16
Total hospitalizations for cancer surgery (any time)
0.5 (0.7) 0.6 (0.7) 0.35 0.4 (0.6) 0.4 (0.6) 0.92 0.2 (0.4) 0.6 (0.6) 0.004 1.0 (0.8) 1.0 (0.6) 0.56
Number of visits within the first six months
5.0 (5.0) 5.7 (5.8) 0.25 4.6 (4.9) 5.3 (6.5) 0.44 5.0 (4.2) 6.6 (6.5) 0.23 6.0 (5.7) 5.8 (3.4) 0.82
Time between diagnosis and date of first treatment(days)

53.3
(39.5)
35.5
(38.1)
<0.001
56.6
(32.5)
42.1
(36.5)
0.005 39.3 (36.1)
22.7
(47.0)
0.12
55.5
(53.0)
33.3
(30.8)
0.019
First intervention 0.40 0.053 0.31 0.67
Surgery
113
(59.5%)
83
(51.9%)

80
(74.8%)
46
(59.0%)
5 (15%)
12
(33%)

28
(56%)
25
(54%)

Radiation 5 (2.6%) 5 (3.1%) 0 (0.0%) 1 (1.3%) 1 (3%) 0 (0%) 4 (8%) 4 (9%)
Drug
therapy
41
(21.6%)
49
(30.6%)

23
(21.5%)
26
(33.3%)
7 (21%) 9 (25%)
11
(22%)
14
(30%)

Palliative
care
26
(13.7%)
20
(12.5%)
2 (1.9%) 5 (6.4%) 18 (55%)
13
(36%)

6 (12%) 2 (4%)
No therapy 5 (2.6%) 3 (1.9%) 2 (1.9%) 0 (0.0%) 2 (6%) 2 (6%) 1 (2%) 1 (2%)
Reason for dose changes 0.94 0.22 0.41
Toxicity 14 (82%) 15 (83%) 5 (100%) 6 (75%) 5 (100%)
5
(100%)

4 (57%) 4 (80%)
Other 3 (18%) 3 (17%) 0 (0%) 2 (25%) 3 (43%) 1 (20%)
Total number of administered cycles in first 90 days of IV therapy
4.4 (2.3) 4.2 (2.3) 0.69 4.8 (2.6) 4.5 (2.5) 0.69 3.1 (1.1) 3.1 (1.7) 0.99 4.6 (2.1) 4.4 (2.2) 0.77
Neoadjuvant therapy 0.02 0.084 0.56 0.13
Yes
34
(18.5%)
44
(29.3%)

19
(18.1%)
21
(29.2%)
5 (15%) 7 (21%)
10
(22%)
16
(36%)

No
150
(81.5%)
106
(70.7%)

86
(81.9%)
51
(70.8%)
28 (85%)
27
(79%)

36
(78%)
28
(64%)

29

All cancers combined Breast cancer Lung cancer Colorectal cancer
Variables* 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value
Adjuvant therapy 0.99 0.75 0.031 0.94
Yes
101
(56.7%)
81
(56.6%)

80
(77.7%)
55
(79.7%)
2 (7%) 9 (27%)
19
(42%)
17
(41%)

No
77
(43.3%)
62
(43.4%)

23
(22.3%)
14
(20.3%)

28
(93%)
24
(73%)

26
(58%)
24
(59%)

Time between diagnosis and date of start of chemotherapy(days)

77.7
(87.1)
67.1
(101.3)
0.43
91.2
(56.3)
67.0
(45.7)
0.021
31.8
(139.7)
88.3
(189.2)
0.29
85.1
(77.7)
51.8
(75.8)
0.083
Time between diagnosis and date of start of radiation therapy(days)

178.2
(187.9)
99.9
(624.4)
0.26
167.7
(85.8)
189.3
(115.1)
0.25
64.6
(117.2)
125.8
(220.2)
0.39
342.8
(404.0)
114.7
(75.8)
0.03
Total imaging scan times
8.8 (8.3) 6.9 (5.3) 0.009 8.1 (6.0) 7.0 (5.0) 0.19
10.8
(10.7)
7.6 (6.3) 0.14
9.3
(10.3)
6.1
(5.1)
0.062
Has patient been infected with COVID-
19?
0.27 0.46 0.72 0.17
Yes 11 (8.3%)
15
(12.5%)
7 (9%) 8 (13%) 2 (12%) 2 (8%)
2 (6%) 5
(16%)

No
122
(91.7%)
105
(87.5%)

73
(91%)
56
(88%)

15
(88%)
22
(92%)

34
(94%)
27
(84%)

Comorbidity 0.085 0.064 0.62 0.38
Hypertension 8 (5.8%) 6 (5.3%) 3 (4%) 4 (7%) 2 (9%) 2 (7%)
3 (8%) 0
(0%)

Hyperlipidemi
a
14
(10.1%)
7 (6.1%) 9 (11%) 3 (5%) 1 (5%) 3 (11%)
4 (11%) 1
(3%)

Osteoarthritis 0 (0.0%) 4 (3.5%) 0 (0%) 3 (11%)
0 (0%) 1
(3%)

Hypothyroidis
m
8 (5.8%) 3 (2.6%) 8 (10%) 2 (4%)
0 (0%) 1
(3%)

Diabetes
mellitus
17
(12.3%)
7 (6.1%)
13
(16%)
3 (5%) 2 (9%) 2 (7%)
2 (6%) 2
(7%)

Coronary
artery disease
3 (2.2%) 2 (1.8%) 1 (1%) 0 (0%) 1 (5%) 1 (4%)
1 (3%) 1
(3%)

Other
88
(63.8%)
85
(74.6%)

46
(57%)
45
(79%)

16
(73%)
16
(59%)

26
(72%)
24
(80%)

Comorbidity (Yes/No) 0.81 0.74 0.35 0.47
Yes
138
(71.5%)
114
(70.4%)
80
(73.4%)
57
(71.3%)
22
(65%)
27
(75%)
36
(72%)
30
(65%)

No
55
(28.5%)
48
(29.6%)
29
(26.6%)
23
(28.7%)
12
(35%)
9 (25%) 14
(28%)
16
(35%)

30

All cancers combined Breast cancer Lung cancer Colorectal cancer
Variables* 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value 2019 2020 P-value
Time between diagnosis and date of surgery(days)
83.5 (67.9) 84.7 (82.3) 0.92
78.9
(93.8)
99.4
(148.8)
0.49
Time between date of surgery and post-surgery mammogram(days)
309.0 (153.4)
271.3(117.9
)
0.15
Time between diagnosis and date of reconstructive surgeries(days)
232.8 (246.2) 85.0 (80.8) 0.051
Time between surgery and date of reconstructive surgery(days)
130.1 (180.6) 15.0 (78.5) 0.042
Type of surgery 0.063
Mastectomy 41 (43%) 38 (58%)
Lumpectom
y
55 (57%) 28 (42%)
Use of neoadjuvant endocrine therapy 0.87
Yes 10 (10.0%) 7 (10.8%)
No 90 (90.0%) 58 (89.2%)
Change of chemotherapy protocol from IV to oral 0.46 0.66
Yes 2 (7%) 1 (3%) 2 (8%) 1 (5%)
No
27
(93%)
33
(97%)
22 (92%) 19 (95%)
Time between surgery and date of actual colonoscopy(days)
77.9
(69.3)
104.5
(178.1)
0.50
Time between surgery and date of radiation(days)
-266.5
(555.7)
97.4
(128.8)
0.075
*Mean and SD are calculated for continuous variables; N and percentage are calculated for categorical variables.

31

Table 3. Univariate determinants of time between diagnosis and treatment
All cancers combined Breast cancer Lung cancer Colorectal cancer
Variables OR (95% CI) * P-value OR (95% CI) * P-value OR (95% CI) * P-value OR (95% CI) * P-value
Site
NCCC (ref) (ref) (ref) (ref)
LAC 2.59 (1.68, 4.01) <0.001 3.16 (1.73, 5.78) <0.001 9.44 (2.41, 36.97) 0.001 1.15 (0.52, 2.58) 0.729
Year
2019 (ref) (ref) (ref) (ref)
2020 0.49 (0.32, 0.75) 0.001 0.41 (0.22, 0.75) 0.004 0..80 (0.31, 2.05) 0.642 0.41 (0.14, 1.20) 0.103
Age 1.00 (0.99, 1.02) 0.625 0.99 (0.97, 1.02) 0.545 0.98 (0.94, 1.02) 0.307 1.03 (1.00, 1.06) 0.046
Sex
Male (ref) (ref) (ref) (ref)
Female 0.56 (0.35, 0.91) 0.019 1.51 (0.134, 16.93) 0.739 0.46 (0.17, 1.20) 0.113 0.41 (0.18, 0.94) 0.036
Race/ethnicity
NHW (ref) (ref) (ref) (ref)
NHB 3.31 (1.12, 9.82) 0.031 5.82 (0.56, 60.531 0.140 1 0.47 (0.08, 2.93) 0.421
Hispanic 1.98 (1.13, 3.46) 0.016 2.09 (1.00, 4.38) 0.048 5.24 (1.05, 25.97) 0.043 1.11 (0.39, 3.16) 0.836
Asian/PI 0.96 (0.52, 1.78) 0.905 0.97 (0.41, 2.28) 0.945 2.00 (0.43, 9.26) 0.375 0.63 (0.19, 2.04) 0.441
Other 1.20 (0.32, 4.60) 0.786 1.16 (0.25, 5.47) 0.847 1 1
Type of Cancer
Breast Cancer (ref)
Lung Cancer 1.20 (0.69, 2.08) 0.516
Colorectal Cancer 1.09 (0.67, 1.79) 0.728
Stage at diagnosis
Stage 0-2 (ref) (ref) (ref) (ref)
Stage 3-4 0.54 (0.35, 0.84) 0.005 0.37 (0.17, 0.81) 0.013 0.74 (0.25, 2.20) 0.583 0.27 (0.11, 0.67) 0.005
Comorbidity
Hypertension (ref) (ref) (ref) (ref)
Hyperlipidemia 1.35 (0.34, 5.44) 0.673 1.79 (0.24, 13.21) 0.570 2.99 (0.15, 59.89) 0.472 0.33 (0.02, 6.65) 0.472
Osteoarthritis 1.80 (0.19, 16.98) 0.608 1.50 (0.06, 40.63) 0.810 1
Hypothyroidism 3.15 (0.61, 16.31) 0.171 5.83 (0.47, 29.75) 0.211 1
Diabetes mellitus 1.29 (0.33, 5.02) 0.718 2.50 (0.37, 16.89) 0.347 1 0.50 (0.22, 11.09) 0.661
Coronary artery disease 1.20 (0.15, 9.776) 0.865 1 2.99 (0.47, 53.68) 0.184 1
Other 1.82 (0.59, 5.65) 0.300 1.88 (0.35, 10.18) 0.345 2.05 (0.19, 21.07) 0.552 0.64 (0.05, 7.48) 0.719
Comorbidity (Yes/NO)
Yes (ref) (ref) (ref) (ref)
No 0.54 (0.34, 0.87) 0.011 0.76 (0.39, 1.45) 0.397 0.59 (0.21, 1.68) 0.323 0.25 (0.10, 0.67) 0.006
(ref) indicates reference group.
*Odds Ratio (OR) and confidence intervals (CI) of the association of each variable with the risk of starting treatment above the median time between diagnosis and treatment in
2019 versus starting treatment at equal or less time than the median in 2019.

32

Table 4. Multivariate determinants of time between diagnosis and first treatment
All cancers combined Breast cancer Lung cancer Colorectal cancer
Determinants OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value
Site
Keck (ref) (ref) (ref)
LAC 2.55 (1.37, 4.75) 0.003 4.18 (2.14, 8.17) <0.001 5.72 (0.83, 39.26) 0.076
Year
2019 (ref) (ref)
2020 0.47 (0.29, 0.77) 0.002 0.34 (0.17, 0.68) 0.002
Age 1.02 (0.99, 1.05) 0.245
Sex
Male (ref) (ref)
Female 0.53 (0.27, 1.06) 0.073 0.39 (0.16, 0.98) 0.046
Race/ethnicity
NHW (ref) (ref)
NHB 1.84 (0.52, 6.52) 0.341 1
Hispanic 1.13 (0.54, 2.34) 0.748 2.39 (0.39, 14.73) 0.347
Asian/PI 0.89 (0.46, 1.73) 0.728 1.86 (0.40, 8.68) 0.429
Other 1.21 (0.28, 5.13) 0.800 1
Type of Cancer
Breast Cancer (ref)
Lung Cancer 1.99 (0.89, 4.45) 0.093
Colorectal Cancer 1.14 (0.56, 2.33) 0.718
Stage at diagnosis
Stage 0-2 (ref) (ref) (ref)
Stage 3-4 0.39 (0.22, 0.69) 0.001 0.41 (0.17, 0.95) 0.038 0.31 (0.12, 0.81) 0.017
(ref) indicates reference group.

33
Table 5. Determinants of first type of treatment for breast cancer patients
Outcome N Determinates OR (95% CI) P-value
Surgery 196
Drug therapy 90 Site
47 Keck (ref)
43 LAC 1.76 (0.57, 5.51) 0.328
Race/ethnicity
17 NHW (ref)
5 NHB 1.40 (0.03, 64.46) 0.864
38 Hispanic 2.07 (0.51, 8.43) 0.306
21 Asian/PI 2.20 (0.61, 7.85) 0.226
3 Other 2.70 (0.32, 22.90) 0.361
Stage at diagnosis
27 Stage 0-2 (ref)
62 Stage 3-4 25.37 (8.55, 75.32) <0.001
Other 64 Site
42 Keck (ref)
22 LAC 0.37 (0.45, 3.03) 0.353
Race/ethnicity
13 NHW (ref)
8 NHB 33.07 (0.41, 2665.24) 0.118
19 Hispanic 3.89 (0.34, 43.79) 0.276
19 Asian/PI 7.56 (0.73, 78.58) 0.090
1 Other 3.81e-06 0.994
Stage at diagnosis
4 Stage 0-2 (ref)
60 Stage 3-4 214.92(19.59, 2358.48) <0.001
(ref) indicates reference group.

Abstract (if available)

Abstract Purpose
The purpose of the study is to investigate the impact of the COVID-19 pandemic on cancer care delivery to identify the most critical changes that may have occurred due to the pandemic, which may have had the largest impact on patient outcomes. We compared the differences in cancer care delivery for patients who started cancer treatment during the COVID-19 pandemic in 2020 with those who started treatment in the same time period in 2019, focusing on identifying key patient-level determinants of specific treatment patterns.
Methods
Using the NCCC and LAC clinical data warehouses, we conducted a chart review of patients who started their treatment for breast, lung, or colorectal cancer at the Norris Comprehensive Cancer Center-Keck Medical Center (NCCC) or the Los Angeles County Medical Center (LAC) from March – September 2019 and those who started treatment in the same period in 2020. A total of 355 patients were included in the analysis. Patient demographic and clinical characteristics, including treatment information, were abstracted from charts. Descriptive statistics was used to examine the distributions of demographics and clinical characteristics of patients, comparing 2019 to 2020 patients. Univariate and multivariate logistic regression were used to examine the potential determinants of time between diagnosis and the date of the first treatment for all cancers and for each type of cancer, considering surgery, drug therapies, and radiation therapies. Polytomous logistic regression was used to identify potential determinants of the type of first intervention for breast cancer patients.
Results
The main difference in patient characteristics was a greater proportion of patients with stage at diagnosis > 3 in 2020 versus 2019, especially among breast cancer patients. When comparing patients who started treatment in 2020 to those in 2019, we observed a higher average number of telemedicine visits (3.8 vs 0.9), a shorter time between diagnosis and date of the first treatment (35.5 days vs 53.3 days), a higher proportion of neoadjuvant therapy (29.3% vs 18.5%), and reduced number of imaging scans (6.9 vs 8.8). In multivariate analyses of time between diagnosis and first treatment, we observed that facility was a key determinant of delays (LAC vs NCCC OR = 2.55; 95% CI: 1.37-4.75, p = 0.003), treatment year and stage and comorbidity were determinants of shorter lag between diagnosis and treatment (2020 vs 2019 OR = 0.47; 95% CI: 0.29-0.77, p = 0.002, stage > 3 vs stage 0-2 OR = 0.39; 95% CI: 0.22-0.69, p = 0.001. Among breast cancer patients, we observed reduced utilization of surgery as first line of treatment in 2020 (59%) compared to 2019 (74.9%), with stage greater than 3 being the key determinant.
Conclusions
Our findings highlight key changes in cancer care delivery and factors associated with care outcomes during the COVID-19 pandemic at the USC Norris Comprehensive Cancer Center and the Los Angeles County Hospital. As expected, the number of telemedicine visits increased significantly in 2020, and surprisingly, the time between diagnosis and the date of the first treatment was shorter in 2020 compared to 2019, likely driven by prioritization of diagnosis and treatment of patients with more advanced cancer. Comparing 2020 to 2019 key observed differences included higher proportion of patients receiving neoadjuvant therapy, reduced imaging scans, and reduced utilization of surgery among breast cancer patients as first line of treatment, with the latter being driven by higher proportion of advanced stage at diagnosis. Overall, the COVID-19 pandemic has had a significant impact on cancer care, and it will be important to continue monitoring its impact on cancer outcomes because of these changes in treatment modalities.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Investigating racial and ethnic disparities in patient experiences with care and health services use following colorectal cancer diagnosis among older adults with comorbid chronic conditions

PDF

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

PDF

Racial/ethnic differences in colorectal cancer patient experiences, health care utilization and their association with mortality: findings from the SEER-CAHPS data

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

An analysis of disease-free survival and overall survival in inflammatory breast cancer

PDF

Disparities in colorectal cancer survival among Latinos in California

PDF

Disparities in gallbladder, intra-hepatic bile duct, and other biliary cancers among multi-ethnic populations: a California Cancer Registry study

PDF

Red and processed meat consumption and colorectal cancer risk: meta-analysis of case-control studies

PDF

Prostate cancer disparities among Californian Latinos by country of origin: clinical characteristics, incidence, treatment received and survival

PDF

Instability of heart rate and rating of perceived exertion during high-intensity interval training in breast cancer patients undergoing anthracycline chemotherapy

PDF

Association between informed decision-making and mental health-related quality of life in long term prostate cancer survivors

PDF

Analysis of factors associated with breast cancer using machine learning techniques

PDF

Associations between isoflavone soy protein (ISP) supplementation and breast cancer in postmenopausal women in the Women’s Isoflavone Soy Health (WISH) clinical trial

PDF

Diet quality and pancreatic cancer incidence in the multiethnic cohort

PDF

Air pollution and breast cancer survival in California teachers: using address histories and individual-level data

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

Fish consumption and risk of colorectal cancer

PDF

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

PDF

Construction of a surgical survival prediction model of stage IV NSCLC patients-based on seer database

PDF

Pathogenic variants in cancer predisposition genes and risk of non-breast multiple primary cancers in breast cancer patients

Asset Metadata

Creator Wu, Yuwei (author)

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

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Degree Conferral Date 2023-05

Publication Date 04/07/2023

Defense Date 04/06/2023

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

Tag cancer care delivery,COVID-19 pandemic,OAI-PMH Harvest

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Stern, Mariana (committee chair), Tsui, Jennifer (committee member), (Farias, Albert)

Creator Email wuyuwei@usc.edu,wuyuwei414659315@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC112956343

Unique identifier UC112956343

Identifier etd-WuYuwei-11584.pdf (filename)

Legacy Identifier etd-WuYuwei-11584

Document Type Thesis

Format theses (aat)

Rights Wu, Yuwei

Internet Media Type application/pdf

Type texts

Source 20230410-usctheses-batch-1018 (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