University of Southern California Dissertations and Theses

Racial and ethnic disparities in delays of surgical treatment for breast cancer

(USC Thesis Other)

Racial and ethnic disparities in delays of surgical treatment for breast cancer

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Racial and Ethnic Disparities in Delays of Surgical Treatment for Breast Cancer

by

Yifei Yang

A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In partial fulfillment of the
Requirement for the Degree
Master of Science
(BIOSTATISTICS)

December 2020

Copyright 2020 Yifei Yang
ii
TABLE OF CONTENTS
List of Tables …………………………………………………………………….………………iii
List of Figures………………………………………………………………………………….…iii
Abstract…………………………………………………………………………………………...iv
Introduction…………………………………………………………………………………...…...1
Methods…………………………………………………………………………………..........…..2
Results…………………………………………………………………………………...………...5
Discussion…………………………………………………………………………………..........16
References………………………………………………………………………………………..22

iii
LIST OF TABLES

Table 1. Baseline characteristics of patients with breast cancer by races/ethnics, diagnosed
between 2012 and 2017 in California……………………………………………………………..7

Table 2. Univariate Analysis for 30-Day Benchmark……………………………………..……11

Table 3. Multivariate Analysis by Logistic Regression for 30-Day Benchmark………....……..13

Table 4. Multivariate Analysis by Cox Proportional Hazard Model………………...………….15

LIST OF FIGURES

Figure 1. Flow chart of sample selection process………………………………………………...4

iv
Abstract
Background: The mortality rate for breast cancer is different across racial/ethnic groups, and
one main contributing factor may be delay in surgery, in addition to biological and
environmental factors. Our study examines the racial and ethnic disparities in delays of breast
cancer surgery.
Methods: We obtained the data from California Cancer Registry (CCR), which has records of
breast cancer cases in California. We included 106,441 females diagnosed with stage I-III
invasive breast cancer between 2012 and 2017 in California. We conducted a multivariable
logistic regression to determine the odds of receiving surgery within 30 days of a diagnosis and a
Cox proportional hazards model to determine the risk of a shorter time to surgery.
Results: Of the study cohort, majority was non-Hispanic White patients (57.5%), followed by
Hispanic patients (20.7%), non-Hispanic black patients (5.9%), Filipino patients (4.3%), Chinese
patients (3.1%), Tahitian patients (2.4%), Japanese patients (1.0%), Asian Indian or Pakistani
patients (1.0%), Korean patients (0.8%) and the rest of the patients were grouped as ‘other’
(3.3%). Compared to white patients, Black and Hispanic patients had lower odds of (Black vs.
White OR: 0.91, 95% CI: 0.85-0.98; Hispanic vs. White OR: 0.94, 95% CI: 0.89-0.98) receiving
surgery within 30 days from diagnosis, and Chinese and Asian Indian or Pakistani patients had a
higher odds (Chinese vs. White OR: 1.29, 95% CI: 1.18-1.41, Asian Indian or Pakistani vs.
White OR: 1.28, 95% CI: 1.11-1.49). Racial ethnic groups that were less likely to receive
surgical treatment within 30 days were also less likely to have a shorter time to surgery
v
comparing to white patients. We found that breast cancer cases diagnosed more recently
experienced a significantly lower odds of receiving surgery within 30 days of diagnosis (2013 vs.
2012 OR: 0.93, 95% CI: 0.88-0.98; 2014 vs. 2012 OR: 0.85, 95% CI: 0.81-0.90; 2015 vs. 2012
OR: 0.76, 95% CI: 0.72-0.81; 2016 vs. 2012 OR: 0.74, 95% CI: 0.70-0.78; 2017 vs. 2012 OR:
0.66, 95% CI: 0.62-0.70).
Discussion: Our study also suggests that there are racial ethnic disparities in time to breast
cancer surgical treatment and that all cases diagnosed more recently experience surgical delays
1
Introduction
In the United States about 12.5% of women will develop breast cancer during their lifetime
1
.
While the overall incidence rate of breast cancer in the United States has stabilized since 2003, in
recent years, it appears the rate has been increasing among racial ethnic groups such as
non-Hispanic Blacks and Asian subgroups including Chinese and Koreans.
2,3
While it is widely
acknowledged that breast cancer incidence and mortality rates tend to be highly associated with
genetics and environmental exposures, unequal cancer treatment also contributes to disparities in
mortality likely because the rising cost of cancer treatment makes high quality care becomes less
affordable and/or accessible.
4
For instance, delay in breast cancer treatment negatively impacts
survival; therefore, it is crucial to investigate additional factors that can lead to a variation in
surgical or system delays.
5
A study by Wright et al showed that after adjusting for certain
demographic factors, Hispanic and Black patients had a higher likelihood of experiencing
surgical delays compared with other race/ethnic groups.
6
However, this study did not control for
patient characteristics such as marital status, type of insurance, and excluded patients that
received neoadjuvant chemotherapy and/or radiation, which are important factors to consider.
Although research has identified various predictors of surgical treatment delay, a gap exists in
exploring the racial/ethnic disparities in surgical treatment with detailed grouping of Asians such
as Korean, Filipino, Japanese and Chinese. Thus, the goal of our study is to characterize
patient-level factors and to evaluate the racial/ethnic disparities detailed on Asian subgroups in
the timespan between breast cancer diagnosis and definitive surgery. We hypothesize that there
2
is a difference in breast cancer surgical delays across racial/ethnic groups.
Methods
Data Source
We conducted a retrospective cohort study for women diagnosed with stage I-III invasive
breast cancer between 2012 and 2017 in California. The data were obtained from the California
Cancer Registry (CCR), a statewide cancer surveillance program that documented the cancer
cases diagnosed in California. The CCR program collects and maintains patient information
including demographic (i.e. racial/ethnic), socioeconomic, clinical, prognostic, treatment and
tumor aspects.
Study Cohort
We included women diagnosed between 2012 and 2017 for stage I-III invasive breast cancer
in California. Figure 1 illustrates our sample selection. We initially started with 116,901
observations, and removed observations for missing values for date of diagnosis or definitive
surgery. We assumed the date to be the 15
th
of the month if only the date was missing for
diagnosis or surgery; otherwise, the observations without both the date and month were excluded.
We then excluded observations where information on having chemotherapy or radiation before
the surgery was unknown. We also excluded observations if they did not have surgery at the
primary site or if the type of surgery was unknown. Lastly, we were left with 106,441
observations for analysis.
To obtain a comprehensive racial/ethnic grouping, we first stratified the data by Hispanic
3
and non-Hispanic. If the observations were non-Hispanic, we further grouped them into more
detailed subgroups for those with at least 800 observations. Thus, we classified the observations
into the following mutually exclusive racial/ethnic groups: non-Hispanic White, non-Hispanic
Black, Hispanic, Chinese, Japanese, Filipino, Korean, Asian Indian or Pakistani, Tahitian and
other.
We abstracted the patient-level covariates from the CCR database. To assess treatment
received we included (1) neoadjuvant chemotherapy, (2) radiation or (3) whether prior to surgery
if they ever received chemotherapy or (4) radiation. For patients reporting never having radiation
or chemotherapy, they were grouped as “not applicable” in the variable of whether they received
chemotherapy or radiation before surgery. Additionally, we examined the following variables:
age at diagnosis, tumor stage, tumor grade, year of diagnosis, metropolitan level, insurance type,
socioeconomic status, marital status, number of positive lymph nodes, primary tumor size, type
of surgical operation, comorbidity status, if ever received chemotherapy or radiation, and if
received chemotherapy or radiation before surgery. In our study, the metropolitan level is
measured by rural-urban commuting area (RUCA): level 1 corresponds to RUCA code of 1,
implying core metropolitan area; level 2 corresponds to RUCA code greater than 1 and less than
7, which is the intermediate level; and level 3 corresponds to RUCA code of 7 or above,
implying small towns and other rural areas. The socioeconomic status refers to the patients’
neighborhood socioeconomic status measured by Juan Yang's index of socioeconomic status,
with level 1 as the lowest SES level and level 5 as the highest SES level.
4

Eligible Participants (n=116901)
Remained Participants (n=111134)
Missing date of diagnosis or surgery
(n=5767)
Remained Participants (n=110382)
Missing date and month of diagnosis
or surgery (n=681)
Remained Participants (n=107957)
Unknown if receiving chemotherapy
before surgery (n=2425)
Unknown if receiving radiation before
surgery (n=511)
Remained Participants (n=107446)
Included in the Analysis (n=106441)
Unknown or none for type of surgery
(n=1005)
Figure 1. Flow chart of sample selection process. Total of 116901 observations obtained from
California Cancer Registry, and 106441 observations left with complete key variables.
5
Statistical Analysis
Prior to running any regression models, we first examined the distribution and prevalence of
the racial/ethnic groups by each potential patient-level covariate. We used Pearson’s chi square
test to see if there was any difference in the levels of covariate across the racial/ethnic group. To
quantify the surgical delays, we created a binary variable to classify patients into either received
surgical treatment within 30 days of diagnosis or not. We then used Pearson’s chi square test the
association between each covariate and the benchmark of 30 days.
As our primary goal is to investigate how surgical delays of breast cancer vary by
racial/ethnic groups, we used multivariable logistic regression models to examine associations
between race/ethnicity and surgical delay controlling for covariates. Additionally, we used Cox
proportional hazards model to assess the duration of time before receiving surgical treatment, a
time-to-event outcome, examined by racial/ethnic groups, so we could investigated the
proportion of the patient groups receiving surgery past a certain time. All analyses were
performed using R statistical software through RStudio Version 1.2.1335 with an alpha level of
0.05.
Results
A total of 106,441 females diagnosed between 2012 and 2017 for stage I-III invasive breast
cancer in California were used for the following analysis. Of the study cohort, majority was
non-Hispanic White patients (57.5%), followed by Hispanic patients (20.7%), non-Hispanic
black patients (5.9%), Filipino patients (4.3%), Chinese patients (3.1%), Tahitian patients (2.4%),
6
Japanese patients (1.0%), Asian Indian or Pakistani patients (1.0%), Korean patients (0.8%) and
the rest of the patients were grouped as ‘other’ (3.3%).
Table 1 shows the distribution of the demographic, socioeconomic, clinical, prognostic,
treatment and tumor characteristics across the racial/ethnic groups. Most of the patients were
aged between 50-69; while very few were below 40 years old or above 80 years. Tumor stage I
and grade II were most common across all the racial/ethnic groups other than Asian Indian or
Pakistani. Year of diagnosis was relatively evenly distributed across the years with slightly
increasing trend over time. Comparing with White patients, all the other racial/ethnic groups
other than Asian Indian or Pakistani had a higher proportion living in the highest metropolitan
level neighborhood. Private insurance was the most common insurance type for all the groups.
For the socioeconomic status, Chinese, Japanese, Korean, Asian Indian or Pakistani and Tahitian
patients had higher percentage in the highest level compared to White patients. Most of the
patients were married, with similar rates across all the racial/ethnic groups; tumor size and
number of lymph nodes demonstrated similar pattern across all the groups as well. Compared to
White patients, all the other groups had a higher percentage of mastectomy as the surgical
treatment. Only Japanese patients (68.0%) had a higher percentage for not receiving
chemotherapy than White patients (66.2%), and only Asian Indian or Pakistani patients (44.0%)
had a lower percentage for not receiving radiation than White patients (46.7%). Korean patients
had the highest percentage for receiving chemotherapy (15.8%) or radiation (0.3%) before

7
Table 1. Baseline characteristics of patients with breast cancer by races/ethnics, diagnosed between 2012 and 2017 in
California.
White Black Hispanic Chinese Japanese Filipino Korean Asian Indian
or Pakistani
Tahitian Other

P value

<40 2234
(3.6%)
375
(6.0%)
1920
(8.7%)
215
(6.5%)
50
(4.9%)
233
(5.1%)
67
(7.5%)
135
(12.3%)
204
(7.9%)
256
(7.3%)
<0.001
40-49 8028
(13.1%)
1016
(16.3%)
5229
(23.7%)
765
(23.2%)
213
(20.7%)
819
(18.1%)
252
(28.1%)
271
(24.7%)
600
(23.2%)
791
(22.4%)
50-59 14261
(23.3%)
1766
(28.3%)
5974
(27.1%)
929
(28.1%)
219
(21.3%)
1232
(27.2%)
252
(28.1%)
256
(23.4%)
710
(27.4%)
918
(26%)
60-69 18439
(30.1%)
1602
(25.7%)
5219
(23.7%)
809
(24.5%)
237
(23.0%)
1371
(30.2%)
188
(21.0%)
258
(23.6%)
648
(25.0%)
974
(27.6%)
70-79 12378
(20.2%)
1061
(17.0%)
2646
(12.0%)
380
(11.5%)
173
(16.8%)
705
(15.5%)
119
(13.3%)
147
(13.4%)
298
(11.5%)
434
(12.3%)
>=80 5879
(9.6%)
410
(6.6%)
1031
(4.7%)
204
(6.2%)
138
(13.4%)
177
(3.9%)
18
(2.0%)
28
(2.6%)
128
(4.9%)
152
(4.3%)
Tumor Stage

I 34339
(56.1%)
2886
(46.3%)
10088
(45.8%)
1726
(52.3%)
616
(59.8%)
2266
(49.9%)
406
(45.3%)
465
(42.5%)
1428
(55.2%)
1768
(50.2%)
<0.001
II 20906
(34.1%)
2481
(39.8%)
8743
(39.7%)
1268
(38.4%)
335
(32.5%)
1739
(38.3%)
396
(44.2%)
484
(44.2%)
942
(36.4%)
1342
(38.1%)
III 5974
(9.8%)
863
(13.9%)
3188
(14.5%)
308
(9.3%)
79
(7.7%)
532
(11.7%)
94
(10.5%)
146
(13.3%)
218
(8.4%)
415
(11.8%)
Tumor Grade

I 16861
(27.5%)
1114
(17.9%)
4401
(20.0%)
702
(21.3%)
295
(28.6%)
891
(19.6%)
156
(17.4%)
235
(21.5%)
596
(23%)
737
(20.9%)
<0.001
II 27409
(44.8%)
2415
(38.8%)
9331
(42.4%)
1501
(45.5%)
444
(43.1%)
2071
(45.6%)
374
(41.7%)
450
(41.1%)
1166
(45.1%)
1551
(44.0%)
III 15350
(25.1%)
2486
(39.9%)
7570
(34.4%)
967
(29.3%)
254
(24.7%)
1419
(31.3%)
331
(36.9%)
366
(33.4%)
728
(28.1%)
1086
(30.8%)
IV 188
(0.3%)
26
(0.4%)
71
(0.3%)
7
(0.2%)
4
(0.4%)
11
(0.2%)
6
(0.7%)
3
(0.3%)
6
(0.2%)
7
(0.2%)
Unknown 1411
(2.3%)
189
(3.0%)
646
(2.9%)
125
(3.8%)
33
(3.2%)
145
(3.2%)
29
(3.2%)
41
(3.7%)
92
(3.6%)
144
(4.1%)
Year of
diagnosis
2012 10163
(16.6%)
1044
(16.8%)
3301
(15.0%)
459
(13.9%)
170
(16.5%)
723
(15.9%)
129
(14.4%)
147
(13.4%)
384
(14.8%)
522
(14.8%)
<0.001
2013 10128
(16.5%)
992
(15.9%)
3417
(15.5%)
511
(15.5%)
184
(17.9%)
734
(16.2%)
118
(13.2%)
163
(14.9%)
359
(13.9%)
553
(15.7%)
2014 10060
(16.4%)
1037
(16.6%)
3444
(15.6%)
515
(15.6%)
158
(15.3%)
704
(15.5%)
139
(15.5%)
165
(15.1%)
391
(15.1%)
540
(15.3%)
2015 10483
(17.1%)
1064
(17.1%)
3730
(16.9%)
591
(17.9%)
172
(16.7%)
803
(17.7%)
154
(17.2%)
207
(18.9%)
455
(17.6%)
589
(16.7%)
2016 10295
(16.8%)
1043
(16.7%)
4086
(18.6%)
606
(18.4%)
172
(16.7%)
758
(16.7%)
173
(19.3%)
198
(18.1%)
480
(18.5%)
647
(18.4%)
2017 10090
(16.5%)
1050
(16.9%)
4041
(18.4%)
620
(18.8%)
174
(16.9%)
815
(18.0%)
183
(20.4%)
215
(19.6%)
519
(20.1%)
674
(19.1%)

8
Table 1. Continued
White Black Hispanic Chinese Japanese Filipino Korean Asian
Indian or
Pakistani
Tahitian Other

P value
Metropolitan
Level
1(lowest) 1626
(2.7%)
33
(0.5%)
276
(1.3%)
5
(0.2%)
4
(0.4%)
14
(0.3%)
1
(0.1%)
2
(0.2%)
6
(0.2%)
73
(2.1%)
<0.001
2 21626
(35.3%)
1501
(24.1%)
6389
(29.0%)
731
(22.1%)
254
(24.7%)
1155
(25.5%)
162
(18.1%)
470
(42.9%)
719
(27.8%)
1035
(29.4%)
3(highest) 37967
(62.0.%)
4696
(75.4%)
15354
(69.7%)
2566
(77.7%)
772
(75.0%)
3368
(74.2%)
733
(81.8%)
623
(56.9%)
1863
(72.0%)
2417
(68.6%)
Insurance Private 36223
(59.2%)
3650
(58.6%)
11760
(53.4%)
2301
(69.7%)
692
(67.2%)
2973
(65.5%)
522
(58.3%)
749
(68.4%)
1779
(68.7%)
1960
(55.6%)
<0.001
Medicare 20040
(32.7%)
1541
(24.7%)
4209
(19.1%)
575
(17.4%)
290
(28.2%)
874
(19.3%)
180
(20.1%)
175
(16.0%)
503
(19.4%)
671
(19.0%)
Medicaid 3024
(4.9%)
789
(12.7%)
5140
(23.3%)
344
(10.4%)
23
(2.2%)
474
(10.4%)
163
(18.2%)
129
(11.8%)
190
(7.3%)
573
(16.3%)
Uninsured 307
(0.5%)
41
(0.7%)
202
(0.9%)
14
(0.4%)
1
(0.1%)
21
(0.5%)
14
(1.6%)
7
(0.6%)
25
(1.0%)
15
(0.4%)
Other 1625
(2.7%)
209
(3.4%)
708
(3.2%)
68
(2.1%)
24
(2.3%)
195
(4.3%)
17
(1.9%)
35
(3.2%)
91
(3.5%)
306
(8.7%)
SES 1(Lowest) 3968
(6.5%)
1451
(23.3%)
6145
(27.9%)
207
(6.3%)
37
(3.6%)
363
(8.0%)
68
(7.6%)
60
(5.5%)
176
(6.8%)
408
(11.6%)
<0.001
2 8856
(14.5%)
1542
(24.8%)
5427
(24.6%)
304
(9.2%)
90
(8.7%)
778
(17.1%)
149
(16.6%)
113
(10.3%)
335
(12.9%)
707
(20.1%)
3 12366
(20.2%)
1361
(21.8%)
4501
(20.4%)
527
(16.0%)
180
(17.5%)
1118
(24.6%)
145
(16.2%)
154
(14.1%)
470
(18.2%)
762
(21.6%)
4 16207
(26.5%)
1198
(19.2%)
3646
(16.6%)
870
(26.3%)
299
(29.0%)
1360
(30.0%)
209
(23.3%)
256
(23.4%)
699
(27.0%)
821
(23.3%)
5(Highest) 19822
(32.4%)
678
(10.9%)
2300
(10.4%)
1394
(42.2%)
424
(41.2%)
918
(20.2%)
325
(36.3%)
512
(46.8%)
908
(35.1%)
827
(23.5%)
Marital status Not married 23488
(38.4%)
3790
(60.8%)
8466
(38.4%)
904
(27.4%)
363
(35.2%)
1428
(31.5%)
249
(27.8%)
202
(18.4%)
767
(29.6%)
1144
(32.5%)
<0.001
Married 34951
(57.1%)
2094
(33.6%)
12540
(57.0%)
2248
(68.1%)
630
(61.2%)
2891
(63.7%)
622
(69.4%)
865
(79.0%)
1674
(64.7%)
2028
(57.5%)
Other
/unknown
2780
(4.5%)
346
(5.6%)
1013
(4.6%)
150
(4.5%)
37
(3.6%)
218
(4.8%)
25
(2.8%)
28
(2.6%)
147
(5.7%)
353
(10.0%)
Positive lymph
nodes
0 42613
(69.6%)
3969
(63.7%)
14025
(63.7%)
2286
(69.2%)
733
(71.2%)
3101
(68.3%)
603
(67.3%)
706
(64.5%)
1851
(71.5%)
2335
(66.2%)
<0.001
1 7531
(12.3%)
885
(14.2%)
2946
(13.4%)
399
(12.1%)
118
(11.5%)
583
(12.8%)
121
(13.5%)
163
(14.9%)
320
(12.4%)
469
(13.3%)
2 or more 7754
(12.7%)
1016
(16.3%)
3971
(18.0%)
416
(12.6%)
114
(11.1%)
646
(14.2%)
131
(14.6%)
179
(16.3%)
287
(11.1%)
510
(14.5%)
Unknown 3321
(5.4%)
360
(5.8%)
1077
(4.9%)
201
(6.1%)
65
(6.3%)
207
(4.6%)
41
(4.6%)
47
(4.3%)
130
(5.0%)
211
(6.0%)

9
Table 1. Continued
White Black Hispanic Chinese Japanese Filipino Korean Asian
Indian or
Pakistani
Tahitian Other

P value
Tumor size <20 24143
(39.4%)
2060
(33.1%)
6885
(31.3%)
1126
(34.1%)
436
(42.3%)
1500
(33.1%)
274
(30.6%)
321
(29.3%)
910
(35.2%)
1113
(31.6%)
<0.001
20-39 11483
(18.8%)
1360
(21.8%)
4651
(21.1%)
676
(20.5%)
171
(16.6%)
975
(21.5%)
177
(19.8%)
238
(21.7%)
465
(18%)
747
(21.2%)
40-59 2714
(4.4%)
373
(6.0%)
1268
(5.8%)
145
(4.4%)
43
(4.2%)
274
(6.0%)
51
(5.7%)
57
(5.2%)
101
(3.9%)
166
(4.7%)
>=60 2494
(4.1%)
344
(5.5%)
1088
(4.9%)
129
(3.9%)
34
(3.3%)
215
(4.7%)
38
(4.2%)
66
(6.0%)
113
(4.4%)
178
(5.0%)
Unknown 20385
(33.3%)
2093
(33.6%)
8127
(36.9%)
1226
(37.1%)
346
(33.6%)
1573
(34.7%)
356
(39.7%)
413
(37.7%)
999
(38.6%)
1321
(37.5%)
Type of
operation
Lumpectomy 39315
(64.2%)
3890
(62.4%)
11984
(54.4%)
1732
(52.5%)
603
(58.5%)
2169
(47.8%)
464
(51.8%)
602
(55.0%)
1356
(52.4%)
1908
(54.1%)
<0.001
Mastectomy 21904
(35.8%)
2340
(37.6%)
10035
(45.6%)
1570
(47.5%)
427
(41.5%)
2368
(52.2%)
432
(48.2%)
493
(45.0%)
1232
(47.6%)
1617
(45.9%)
Chemotherapy No 40531
(66.2%)
3452
(55.4%)
11743
(53.3%)
1966
(59.5%)
700
(68%)
2538
(55.9%)
483
(53.9%)
550
(50.2%)
1626
(62.8%)
2066
(58.6%)
<0.001
Yes 20688
(33.8%)
2778
(44.6%)
10276
(46.7%)
1336
(40.5%)
330
(32%)
1999
(44.1%)
413
(46.1%)
545
(49.8%)
962
(37.2%)
1459
(41.4%)
Radiation No 28580
(46.7%)
3251
(52.2%)
11750
(53.4%)
1593
(48.2%)
520
(50.5%)
2458
(54.2%)
495
(55.2%)
482
(44.0%)
1457
(56.3%)
1918
(54.4%)
<0.001
Yes 32639
(53.3%)
2979
(47.8%)
10269
(46.6%)
1709
(51.8%)
510
(49.5%)
2079
(45.8%)
401
(44.8%)
613
(56.0%)
1131
(43.7%)
1607
(45.6%)
Chemotherapy
before surgery
No 15831
(25.9%)
2054
(33.0%)
7310
(33.2%)
1022
(31.0%)
266
(25.8%)
1532
(33.8%)
271
(30.2%)
382
(34.9%)
703
(27.2%)
1087
(30.8%)
<0.001
Yes 4857
(7.9%)
724
(11.6%)
2966
(13.5%)
314
(9.5%)
64
(6.2%)
467
(10.3%)
142
(15.8%)
163
(14.9%)
259
(10.0%)
372
(10.6%)
Not
applicable
40531
(66.2%)
3452
(55.4%)
11743
(53.3%)
1966
(59.5%)
700
(68.0%)
2538
(55.9%)
483
(53.9%)
550
(50.2%)
1626
(62.8%)
2066
(58.6%)
Radiation
before surgery
No 32557
(53.2%)
2966
(47.6%)
10230
(46.5%)
1706
(51.7%)
510
(49.5%)
2073
(45.7%)
398
(44.4%)
611
(55.8%)
1128
(43.6%)
1604
(45.5%)
<0.001
Yes 82
(0.1%)
13
(0.2%)
39
(0.2%)
3
(0.1%)
0
(0.0%)
6
(0.1%)
3
(0.3%)
2
(0.2%)
3
(0.1%)
3
(0.1%)
Not
applicable
28580
(46.7%)
3251
(52.2%)
11750
(53.4%)
1593
(48.2%)
520
(50.5%)
2458
(54.2%)
495
(55.2%)
482
(44.0%)
1457
(56.3%)
1918
(54.4%)
Comorbidity
Status
0 35608
(58.2%)
2873
(46.1%)
10916
(49.6%)
1930
(58.4%)
601
(58.3%)
2041
(45.0%)
550
(61.4%)
551
(50.3%)
1373
(53.1%)
1867
(53.0%)
<0.001
1 8644
(14.1%)
1146
(18.4%)
3428
(15.6%)
296
(9.0%)
141
(13.7%)
777
(17.01%)
85
(9.5%)
159
(14.5%)
339
(13.1%)
443
(12.6%)
2 or more 4742
(7.7%)
942
(15.1%)
1743
(7.9%)
125
(3.8%)
72
(7.0%)
410
(9.0%)
26
(2.9%)
75
(6.8%)
140
(5.4%)
232
(6.6%)
Unknown 12225
(20.0%)
1269
(20.4%)
5932
(26.9%)
951
(28.8%)
216
(21.0%)
1309
(28.9%)
235
(26.2%)
310
(28.3%)
736
(28.4%)
983
(27.9%)

10
surgery, and Japanese patients had the lowest percentage for receiving chemotherapy (6.2%) or
radiation (0.0%) before surgery. Comorbidity status measured by Charlson Comorbidity Index
indicated that Black patients had the highest percentage for the highest-level comorbid
conditions (15.1%), whereas Korean patients had the lowest percentage (2.9%).
Surgical treatment 30-day benchmark and demographic, socioeconomic, clinical, prognostic,
treatment and tumor characteristics are shown in Table 2. Overall, the majority of patients did
not received surgical treatment within 30 days of the diagnosis. Asian subgroups and White
patients had higher proportion of having surgical treatment within 30 days of the diagnosis,
compared to the overall population percentage at 18.5%. Hispanic patients had the lowest
proportion, at 16.9% of having surgical treatment within 30 days of the diagnosis. Globally
across the population, patients aged between 40 to 69 years old had more surgical delays. Tumor
stage of III and grade II were more prone to treatment delay. Interestingly among the entire
population based on 30-day benchmark, in 2012, 19% reported no surgical delays while 15.3%
reported delays, it became 14.5% and 17.9% respectively in 2017, suggesting that surgical delays
increased over the years. Patients with low level of metropolitan level and private insurance had
less surgical delays. Having higher socioeconomic status appears to reduce surgical delays: 27.9%
received treatment within 30 days; whereas at the lowest level, only 10.9% received treatment
within 30 days. Unmarried patients had more surgical delays compared with married patients.
With increasing number of positive lymph nodes, surgical delays also decreased, but smaller
tumor size (<20mm) was associated with less surgical delays. Mastectomy, a more intense breast
11
Table 2. Univariate Analysis for 30-Day Benchmark

≤ 30 days

> 30 days P value
Age group <40 1185 6.0% 4504 5.2% <0.001
40-49 3236 16.4% 14748 17.0%
50-59 4883 24.8% 21634 24.9%
60-69 5346 27.1% 24399 28.1%
70-79 3441 17.5% 14900 17.2%
>=80 1601 8.1% 6564 7.6%
Tumor Stage

I 10253 52.1% 45735 52.7% <0.001
II 6872 34.9% 31764 36.6%
III 2567 13.0% 9250 10.7%
Tumor Grade I 4898 24.9% 21090 24.3% <0.001
II 7965 40.4% 38747 44.7%
III 5933 30.1% 24624 28.4%
IV 85 0.4% 244 0.3%
Unknown 811 4.1% 2044 2.4%
Year of diagnosis 2012 3744 19.0% 13298 15.3% <0.001
2013 3555 18.1% 13604 15.7%
2014 3292 16.7% 13861 16.0%
2015 3160 16.0% 15088 17.4%
2016 3081 15.6% 15377 17.7%
2017 2860 14.5% 15521 17.9%
Metropolitan Level 1(lowest) 423 2.1% 1617 1.9% <0.001
2 6986 35.5% 27056 31.2%
3(highest) 12283 62.4% 58076 66.9%
Insurance Private 12139 61.6% 50470 58.2% <0.001
Medicare 5115 26.0% 23943 27.6%
Medicaid 1682 8.5% 9167 10.6%
Uninsured 125 0.6% 522 0.6%
Other 631 3.2% 2647 3.1%
SES 1(Lowest) 2145 10.9% 10738 12.4% <0.001
2 3300 16.8% 15001 17.3%
3 3985 20.2% 17599 20.3%
4 4777 24.3% 20788 24.0%
5(Highest) 5485 27.9% 22623 26.1%
Marital status Not married 7061 35.9% 33740 38.9% <0.001
Married 11569 58.7% 48974 56.5%
Other/unknown 1062 5.4% 4035 4.7%
number of positive lymph
nodes
0 12382 62.9% 59840 69.0% <0.001
1 2427 12.3% 11108 12.8%
2 or more 3281 16.7% 11743 13.5%
Unknown 1602 8.1% 4058 4.7%
Tumor size <20 7689 39.0% 31079 35.8% <0.001
20-39 4051 20.6% 16892 19.5%
40-59 1022 5.2% 4170 4.8%
>=60 989 5.0% 3710 4.3%
Unknown 5941 30.2% 30898 35.6%
Type of operation Lumpectomy 12543 63.7% 51480 59.3% <0.001
Mastectomy 7149 36.3% 35269 40.7%
Chemotherapy No 12089 61.4% 53566 61.7% <0.001
Yes 7603 38.6% 33183 38.3%
Radiation No 9140 46.4% 43364 50.0% <0.001
Yes 10552 53.6% 43385 50.0%
Chemotherapy before
surgery
No 7579 38.5% 22879 26.4% <0.001
Yes 24 0.1% 10304 11.9%
Not applicable 12089 61.4% 53566 61.7%
Radiation before surgery No 10543 53.5% 43240 49.8% <0.001
Yes 9 0.0% 145 0.2%
Not applicable 9140 46.4% 43364 50.0%
12
Table 2. Continued
cancer surgery was associated with more surgical delays than lumpectomy. Patients who had
chemotherapy (38.5% vs. 26.4%) or radiation (53.5% vs. 49.6%) before surgery tended not to
receive treatment within 30 days.
Table 3 shows the multivariable logistic regression analyses. Adjusting for patients’
characteristics, compared to White patients, Chinese patients (OR 1.29, 1.18 - 1.41) and Asian
Indian or Pakistani patients (OR 1.28, 1.11 - 1.49) had a higher odds of receiving surgical
treatment within 30 days of breast cancer diagnosis, while Black patients (OR 0.91, 0.85 - 0.98)
and Hispanic patients (OR 0.94, 0.89 - 0.98) had lower odds. We did not find significant
difference for Japanese, Filipino, Korean, Tahitian and other patients. There was an increasing
likelihood of receiving treatment within 30 days among patients with high tumor stage and grade,
controlling for the other covariate factors, as well as for patients with earlier year of diagnosis,
lower level of metropolitan level, private insurance, higher socioeconomic status, and more
positive number of lymph nodes. Also, married patients (OR 1.10, 1.06 - 1.14) had high odds of
receiving treatment in 30 days, and all the age groups above 40 years old had much lower odds
≤ 30 days > 30 days P value
Comorbidity
Status
0 10563 53.6% 47747 55.0% <0.001
1 2827 14.4% 12631 14.6%
2 or more 1521 7.7% 6986 8.1%
unknown 4781 24.3% 19385 22.3%
Race/Ethnicity White 11587 58.8% 49632 57.2% <0.001
Black 1061 5.4% 5169 6.0%
Hispanic 3732 19.0% 18287 21.1%
Chinese 752 3.8% 2550 2.9%
Japanese 189 1.0% 841 1.0%
Filipino 805 4.1% 3732 4.3%
Korean 170 0.9% 726 0.8%
Asian Indian or
Pakistani
Tahitian
Other
252

440
704
1.3%

2.2%
3.6%
843

2148
2821
1.0%

2.5%
3.3%

13
Table 3. Multivariate Analysis by Logistic Regression for 30-Day Benchmark

≤ 30 days OR (95%CI)

Age group <40 1
40-49 0.69 0.64 - 0.75
50-59 0.69 0.64 - 0.75
60-69 0.68 0.63 - 0.74
70-79 0.73 0.67 - 0.80
>=80 0.67 0.61 - 0.74
Tumor Stage

I 1
II 1.10 1.05 - 1.14
III 1.70 1.57 - 1.84
Tumor Grade I 1
II 0.88 0.85 - 0.92
III 1.09 1.04 - 1.14
IV 1.51 1.16 - 1.97
Unknown 1.98 1.80 - 2.17
Year of diagnosis 2012 1
2013 0.93 0.88 - 0.98
2014 0.85 0.81 - 0.90
2015 0.76 0.72 - 0.81
2016 0.74 0.70 - 0.78
2017 0.66 0.62 - 0.70
Metropolitan Level 1(lowest) 1
2 0.95 0.85 - 1.06
3(highest) 0.79 0.71 - 0.89
Insurance Private 1
Medicare 0.88 0.84 - 0.92
Medicaid 0.83 0.78 - 0.88
Uninsured 0.99 0.81 - 1.22
Other 0.90 0.82 - 0.99
SES 1(Lowest) 1
2 1.05 0.98 - 1.11
3 1.05 0.99 - 1.11
4 1.05 0.99 - 1.12
5(Highest) 1.09 1.03 - 1.16
Marital status Not married 1
Married 1.10 1.06 - 1.14
Other/unknown 1.19 1.10 - 1.29
number of positive lymph nodes 0 1
1 0.98 0.93 - 1.03
2 or more 1.08 1.01 - 1.16
Unknown 2.23 2.08 - 2.39
Type of operation Lumpectomy 1
Mastectomy 0.85 0.81 - 0.88
Chemotherapy before surgery No 1
Yes 0.01 0.00 - 0.01
Not applicable 0.78 0.75 - 0.82
Radiation before surgery No 1
Yes 0.41 0.21 - 0.83
Not applicable 0.87 0.84 - 0.90
Comorbidity
Status
0 1
1 1.03 0.99 - 1.08
2 or more 0.99 0.93 - 1.05
Unknown 1.21 1.16 - 1.26
14
Table 3. Continued

around 0.7 of receiving treatment. For patients who had chemotherapy (OR 0.01, 0.00 - 0.01) or
radiation (OR 0.41, 0.21-0.83) before surgery, they had surgical delays. We did not find any
statistically significant difference in surgical delay across different comorbidity status.
In Table 3, we omitted the variable tumor size in the logistic regression model due to
collinearity, as it is reasonable that cancer stage takes into the account of the tumor size. In
addition, due to timespan of whether patients have ever received chemotherapy or radiation
beyond the 30 days benchmark from diagnosis to treatment, we categorize the patients’ ‘no’ for
this group as ‘not applicable’ if they received the surgery within 30 days. This assumption also
applied for the cox proportional hazard model in Table 4.
Table 4 shows the hazard risk of receiving surgical treatment, a time-to-event outcome,
differed by racial/ethnic groups. Instead of looking at the 30 days benchmark, we focused on the
time interval from breast cancer diagnosis to definitive surgery and examined how the time to
surgery varied across the racial/ethnic groups. The number of days from diagnosis to surgery
overall ranged from 1 to 50175, with a mean of 507 and median of 108. We found that compared
to White patients, Korean patients were more likely to receive surgery (HR=1.10, 1.03 - 1.18),
<= 30 days OR (95%CI)

Race/Ethnicity White 1
Black 0.91 0.85 - 0.98
Hispanic 0.94 0.89 - 0.98
Chinese 1.29 1.18 - 1.41
Japanese 0.95 0.81 - 1.12
Filipino 0.95 0.87 - 1.03
Korean 1.16 0.97 - 1.38
Asian Indian or Pakistani 1.28 1.11 - 1.49
Tahitian 0.91 0.82 - 1.01
Other 1.08 0.99 - 1.18
15
Table 4. Multivariate Analysis by Cox Proportional Hazard Model

HR (95%CI)

Age group <40 1
40-49 0.88 0.86 - 0.91
50-59 0.87 0.85 - 0.90
60-69 0.91 0.88 - 0.94
70-79 0.99 0.95 - 1.02
>=80 1.02 0.98 - 1.06
Tumor Stage

I 1
II 0.95 0.94 - 0.97
III 0.96 0.94 - 0.99
Tumor Grade I 1
II 0.98 0.96 - 0.99
III 1.06 1.04 - 1.08
IV 1.08 0.97 - 1.20
Unknown 1.11 1.07 - 1.16
Year of diagnosis 2012 1
2013 0.98 0.96 – 1.00
2014 0.91 0.89 - 0.93
2015 0.88 0.87 - 0.90
2016 0.85 0.84 - 0.87
2017 0.87 0.85 - 0.89
Metropolitan Level 1(lowest) 1
2 1.00 0.95 - 1.04
3(highest) 0.92 0.88 - 0.96
Insurance Private 1
Medicare 0.87 0.86 - 0.89
Medicaid 0.77 0.75 - 0.79
Uninsured 0.90 0.83 - 0.97
Other 0.86 0.83 - 0.89
SES 1(Lowest) 1
2 1.03 1.00 - 1.05
3 1.05 1.03 - 1.08
4 1.10 1.07 - 1.12
5(Highest) 1.10 1.08 - 1.13
Marital status Not married 1
Married 1.08 1.07 - 1.10
Other/unknown 1.03 1.00 - 1.06
number of positive lymph nodes 0 1
1 0.98 0.96 – 1.00
2 or more 0.99 0.97 - 1.02
Unknown 1.02 0.99 - 1.05
Type of operation Lumpectomy 1
Mastectomy 0.91 0.90 - 0.93
Chemotherapy before surgery No 1
Yes 0.20 0.19 - 0.20
Not applicable 0.78 0.76 - 0.79
Radiation before surgery No 1
Yes 0.55 0.47 - 0.65
Not applicable 0.87 0.86 - 0.89
Comorbidity
Status
0 1
1 1.00 0.98 - 1.02
2 or more 0.98 0.96 - 1.01
Unknown 0.86 0.85 - 0.87
16
Table 4. Continued
and patients in the other racial/ethnic group were less likely to receive surgery (HR=0.92, 0.89 -
0.96). For tumor stage, stage II and III were associated with receiving surgery for patients
(HR=0.95 and 0.96).
In addition, because we manually set the missing date to be the 15
th
of month for diagnosis
and surgery, when there is negative time interval, it meant that the interval between their surgery
date and diagnosis were very short, ranging from 0 days to 14 days. Thus, in this case we took
the average and recoded them as 7 days when running the Cox proportional hazard model. We
also performed sensitivity analysis by setting the time intervals to 0 or 14 days, but it did not
change the overall results and findings were similar to 7 days.
Discussion
Surgical delays are known to negatively impact the survival outcomes of breast cancer.
5
In
this retrospective cohort study based on the California Cancer Registry data, we examined the
disparities in surgical delays between different racial/ethnic groups controlling for demographic,
socioeconomic, clinical, prognostic, treatment and tumor characteristics. Our univariate analysis
using Pearson’s chi square test suggests these characteristics varied across racial/ethnic groups.

HR (95%CI)
Race/Ethnicity
White 1
Black 0.92 0.89 - 0.94
Hispanic 0.95 0.94 - 0.97
Chinese 1.15 1.11 - 1.19
Japanese 0.96 0.91 - 1.03
Filipino 0.99 0.96 - 1.02
Korean 1.10 1.03 - 1.18
Asian Indian or Pakistani
Tahitian
Other
1.10
0.96
0.92
1.03 - 1.17
0.92 - 1.00
0.89 – 0.96
17
We further examined the association with logistic regression and cox proportional hazard models.
Adjusting for patient-level covariates and comparing to White patients, multivariable logistic
regression analyses using the 30-day benchmark, we found that Black and Hispanic patients had
statistically significant surgical delays, while Chinese and Asian Indian or Pakistani patients had
less surgical delay. The results for Black and Hispanic patients are consistent with a study by
Wright et al on 30-day benchmark, but we were able to examine the disparities across more
race/ethnic groups, in particular among Asian subgroups.
6
Based on the time from diagnosis to
treatment by cox proportional hazard model, the overall results were similar in that for the
racial/ethnic groups that did not receive surgery within 30 days also tended to have a lower
hazard risk of getting surgery. We found that Korean patients had a higher risk of receiving
surgery and patients in the other racial/ethnic group had a lower risk of receiving surgery. Since
women in the other racial/ethnic group is a mixture of the small number of observations, it is
difficult to draw conclusions to this particular group.

A previous study suggested that breast cancer surgery delay is associated with older age, and
our study demonstrated that after age of 40 years old, the odds of surgery delay is significant and
remains constant afterwards.
7
Our study’s finding is also consistent with Plotogea et al that
higher stage at diagnosis is associated with higher likelihood of receiving definitive treatment
faster.
8
However, the hazard risk by our Cox proportional hazard model showed that patients
with higher tumor stage had less likelihood of receiving surgery faster than that of at the lower
stage. The potential reason could be that we manually assign the dates of surgery or diagnosis as
18
well as the days between diagnosis and surgery for some observations missing an exact date of
surgery or diagnosis. We need to do more examination of the data to explain this finding.
We also see a statistically significant surgical delay comparing the patient neighborhood
metropolitan level 3 to level 1, it is probably due to more barriers to care in small towns or rural
areas. Given that there was no significant for level 2 compared to level 1, future analysis by
spatial analysis using neighborhood longitudes and latitudes would be helpful to examine
whether distance to health care facilities serve as barrier to health care.
Compared to Medicare and Medicaid, patients with private insurance had significantly less
surgical delays. This result is consistent with many other breast cancer studies, as well as
findings from studies examining other dieseases.
9,10
Different type of insurance is associated with
accessibility and affordability of surgical treatment, thereby affecting the time to surgery. We did
not observe significant difference between patients with private insurance and no insurance, but
the study by Eryln et al. suggested that time to treatment are longer for the patients without
insurance.
11
In our study, our results could be due to the small sample size of patients without
insurance.
We also found that patients undergoing mastectomy were more likely to have surgical delays
than women receiving lumpectomy. In addition to the cost difference between the two different
type of surgery and nature of the tumor characteristics, it’s possible that patients choosing
mastectomy were taking more time with breast conserving consultation including plastic surgery
reconstruction or second opinion.
6
However, we do not have any evidence in our data to evaluate
19
this.
Patients that received chemotherapy or radiation before surgery had a higher likelihood of
surgery delay. When we conducted additional logistic regression testing the interaction between
race/ethnic groups and the status of whether patients received chemotherapy, we concluded that
receiving chemotherapy before surgery did not differentially influence the association between
race/ethnicity and time to surgical treatment. This again confirmed the validity of our model in
Table 3.
Interestingly, our study showed that breast cancer surgical delays either within 30 days or
overall increased over the years, from 2012 to 2017, in the multivariable logistic regression
analysis and Cox proportional hazard model. Our further analysis suggested that the proportion
of patients undergoing chemotherapy before surgery also increased over the years. Patients were
not likely to receive surgery treatment within 30 days if they had chemotherapy before. There
was no clear trend in proportion of having radiation before surgery, type of insurance and cancer
stage between 2012 and 2017 in our sample. Many recent studies have suggested that by having
chemotherapy before surgery would increase the chance of having lumpectomy instead of
mastectomy, chemotherapy are more extensively used for more aggressive cancer stages which
usually require the operation of mastectomy.
12

Our study has some limitations. Although we controlled for patient-level covariates, we did
not include facility-level covariates such as hospital surgical volume. For future analysis, we can
include key hospital-related factors from Office of Statewide Health Planning and Development
20
(OSHPD) to examine the characteristics of the hospital where they perform the surgery. Also,
there could be other individual patient factors such as different preferences when seeking care or
cultural implications; yet we are not able to take into account of this in our study.
Despite controlling for many patient-level factors which had great significant association in
surgical delay shown in our study, we found that racial/ethnic disparities still exist, such that the
time from diagnosis to treatment is greatly impacted. As the time to treatment for breast cancer is
also significantly associated with the disease-specific survival, it is essential to address these
disparities and potentially alleviate such gap in access to surgical treatment.
13
For every 30 days
delay to surgery, patients with breast cancer had a 9% increase in mortality for all racial and
ethnic groups.
13
As our study findings suggest that Black and Hispanic patients tend to have
more surgical delays, it implies that delays can be associated with higher breast cancer specific
mortality rate. This has previously been found in other studies, where compared to White
patients, Black and Hispanic patients are more likely to die from breast cancer.
14,15

In conclusion, our study suggests there are significant disparities in surgical delay across
racial/ethnic groups after controlling for many patient-level factors. Therefore, there is an urgent
need for implementation of equal access of quality care. In this study, we also characterize and
examine demographic, socioeconomic, clinical, prognostic, treatment and tumor aspects that can
significantly impact the time to surgery. There is a need for more studies to address and quantify
the appropriate timeframe from treatment to diagnosis. For example, different factors that
contribute to the different lengths of delays, such as addressing the higher tumor stage at
21
diagnosis tend to get surgery more within 30 days could be important, but less likely to have a
shorter time to surgery in the longer term.

22
References
1. DeSantis, Carol, et al. "Breast cancer statistics, 2013." CA: a cancer journal for
clinicians 64.1 (2014): 52-62.
2. Hou, Ningqi, and Dezheng Huo. "A trend analysis of breast cancer incidence rates in the
United States from 2000 to 2009 shows a recent increase." Breast cancer research and
treatment 138.2 (2013): 633-641.
3. Gomez, Scarlett Lin, et al. "Cancer incidence trends among Asian American populations
in the United States, 1990–2008." Journal of the National Cancer Institute 105.15 (2013):
1096-1110.
4. Institute of Medicine. Delivering High-Quality Cancer Care: Charting a New Course for a
System in Crisis. In. Washington, DC: The National Academies Press; 2013.
5. Richards, M. A., et al. "The influence on survival of delay in the presentation and
treatment of symptomatic breast cancer." British journal of cancer 79.5 (1999): 858-864.
6. Wright, G. Paul, et al. "Time from diagnosis to surgical treatment of breast cancer: factors
influencing delays in initiating treatment." The American Surgeon 76.10 (2010):
1119-1122.
7. Wagner, Jamie L., et al. "Delays in primary surgical treatment are not associated with
significant tumor size progression in breast cancer patients." Annals of surgery 254.1
(2011): 119.
8. Caplan, Lee. "Delay in breast cancer: implications for stage at diagnosis and
survival." Frontiers in public health 2 (2014): 87.
9. Hoppe, Ethan J., et al. "Racial disparities in breast cancer persist despite early detection:
analysis of treatment of stage 1 breast cancer and effect of insurance status on
disparities." Breast cancer research and treatment 173.3 (2019): 597-602.
10. Adamson, Adewole S., et al. "Association of delays in surgery for melanoma with
insurance type." JAMA dermatology 153.11 (2017): 1106-1113.
11. Smith, Erlyn C., Argyrios Ziogas, and Hoda Anton-Culver. "Delay in surgical treatment
and survival after breast cancer diagnosis in young women by race/ethnicity." JAMA
surgery 148.6 (2013): 516-523.
12. “Chemotherapy Before Surgery Increases Likelihood of Lumpectomy.” Breastcancer.org,
5 Mar. 2015,
www.breastcancer.org/research-news/lumpectomy-after-neoadjuvant-chemo.
13. Bleicher, Richard J., et al. "Time to surgery and breast cancer survival in the United
States." JAMA oncology 2.3 (2016): 330-339.
14. Banegas, Matthew P., and Christopher I. Li. "Breast cancer characteristics and outcomes
among Hispanic Black and Hispanic White women." Breast cancer research and
treatment 134.3 (2012): 1297-1304.
23
15. Tian, Nancy, et al. "Identifying risk factors for disparities in breast cancer mortality
among African-American and Hispanic women." Women's Health Issues 22.3 (2012):
e267-e276.

Abstract (if available)

Abstract Background: The mortality rate for breast cancer is different across racial/ethnic groups, and one main contributing factor may be delay in surgery, in addition to biological and environmental factors. Our study examines the racial and ethnic disparities in delays of breast cancer surgery. ❧ Methods: We obtained the data from California Cancer Registry (CCR), which has records of breast cancer cases in California. We included 106,441 females diagnosed with stage I-III invasive breast cancer between 2012 and 2017 in California. We conducted a multivariable logistic regression to determine the odds of receiving surgery within 30 days of a diagnosis and a Cox proportional hazards model to determine the risk of a shorter time to surgery. ❧ Results: Of the study cohort, majority was non-Hispanic White patients (57.5%), followed by Hispanic patients (20.7%), non-Hispanic black patients (5.9%), Filipino patients (4.3%), Chinese patients (3.1%), Tahitian patients (2.4%), Japanese patients (1.0%), Asian Indian or Pakistani patients (1.0%), Korean patients (0.8%) and the rest of the patients were grouped as ‘other’ (3.3%). Compared to white patients, Black and Hispanic patients had lower odds of (Black vs. White OR: 0.91, 95% CI: 0.85-0.98

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

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

Disparities in colorectal cancer survival among Latinos in California

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

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

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

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

PDF

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

PDF

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

PDF

Comparison of Cox regression and machine learning methods for survival analysis of prostate cancer

PDF

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

PDF

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

PDF

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

PDF

Analysis of factors associated with breast cancer using machine learning techniques

PDF

Machine learning-based breast cancer survival prediction

PDF

Transposable element suppression in basal-like breast cancer

PDF

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

PDF

Survival trends and related outcomes of survivors of childhood and young adult cancer

PDF

Comparative study of the POG and FNCLCC grading systems in non-rhabdomyosarcoma soft tissue sarcomas

PDF

Screening and association testing of coding variation in steroid hormone coactivator and corepressor genes in relationship with breast cancer risk in multiple populations

PDF

Racial/ethnic diversity and disparities in the child welfare system: a multilevel perspective

Asset Metadata

Creator Yang, Yifei (author)

Core Title Racial and ethnic disparities in delays of surgical treatment for breast cancer

School Keck School of Medicine

Degree Master of Science

Degree Program Biostatistics

Publication Date 12/12/2020

Defense Date 12/10/2020

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

Tag breast cancer,ethnic disparity,OAI-PMH Harvest,racial disparity,surgical delay

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Farias, Albert (committee chair), Kim, Sue E. (committee member), Liu, Lihua (committee member)

Creator Email sunnyyang1994@gmail.com,yangyife@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-403797

Unique identifier UC11668654

Identifier etd-YangYifei-9208.pdf (filename),usctheses-c89-403797 (legacy record id)

Legacy Identifier etd-YangYifei-9208.pdf

Dmrecord 403797

Document Type Thesis

Rights Yang, Yifei

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

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