












































 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e1                                              Cancer Health Disparities 

RESEARCH 

Cervical cancer outcome by type of 
health care facilities: National Cancer 
Database, 2004-2015 
HyounKyoung G. Park1,4, Zhixin E. Wang1,4, Chenguang Wang2, Warner K. Huh3,4, Sejong Bae1,4 

1Division of Preventive Medicine, Department of Medicine, School of Medicine, University of 
Alabama at Birmingham, Birmingham, AL 
2Division of Biostatistics and Bioinformatics, Sidney Kimmel Comprehensive Cancer Center, 
The Johns Hopkins University, Baltimore, MD 
3Division of Gynecology Oncology, Department of Obstetrics and Gynecology, School of 
Medicine, University of Alabama at Birmingham, Birmingham, AL 
4O’Neal Comprehensive Cancer Center, University of Alabama at Birmingham, Birmingham, AL 
 

ABSTRACT 
The National Cancer Database from 2004 to 2015 was analyzed to identify cervical cancer outcomes 
associated with demographic and clinical characteristics measured by types of facility. Chi-Square tests 
were used to compare proportions and logistic regression to determine factors associated with cervical 
cancer outcomes. Women treated at Academic/Research Programs (ARPs) were younger at diagnosis, 
more likely black, less educated and more in Stage 2, lived further away from treatment facilities, had 
less comorbidities and better 5- year survival, and were more likely to be alive at 30 and 90 days after 
surgery compared to other programs. Women treated at Community Cancer Programs were more 
likely 75 and older at diagnosis, more likely to receive radiation treatment and more in Stage 4, more 
living in rural areas and less than 10 miles from the facility, and had more comorbidities, and lower 5-
year survival compared to other programs. Women treated at Comprehensive Community Cancer 
Programs were more likely white and educated, had more private insurance, and underwent surgery. 
Women treated at Integrated Network Cancer Programs were more likely to live in urban, south region, 
and in Stage 1B2, had more surgery and one comorbidity, and died fewer than 30 days after surgery. 
The type of facility and treatment had varied effects on mortality and 5-year survival. Considering the 
different cervical cancer outcomes from different health care facilities, further research is needed to 
identify what factors influence women to choose a health care facility for their treatment and how this 
choice can affect different health outcomes. 

KEYWORDS: Cervical cancer, type of facility, mortality, survival 

Citation: Park HG et al (2019) Cervical cancer outcome by type of health care facilities: National Cancer 
Database, 2004-2015; Cancer Health Disparities. 4: e1-e-13. doi:10.9777/chd.2019.1011



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e2                                              Cancer Health Disparities 

RESEARCH 

Introduction 

Cervical cancer is the most common type of 
gynecological cancer but one of the most 
preventable and treatable cancers (Fedewa et al., 
2012; Ferlay et al., 2015). In the United States, 
cervical cancer incidence and mortality rates 
continue to decline since regular cervical cancer 
screening tests and vaccination were introduced 
(Akinlotan et al., 2017; CDC, 2018; Scarinci et al., 
2010.). 

Several studies have investigated risk factors 
associated with higher cervical cancer incidence 
and mortality. For example, being black (Arvizo 
and Mahdi, 2017; Beavis et al., 2017; Furlow, 2018; 
Powell et al., 2018 ; Scarinci et al., 2010. ; Singh and 
Jemal, 2017; Weragoda et al., 2016 ; Yoo et al., 
2017), getting older (Fedewa et al., 2012; Furlow, 
2018; Yosta and Hoekstra, 2018), living in The 
South (Yoo et al., 2017), less income (Singh and 
Jemal, 2017), less education (Singh and Jemal, 
2017), having comorbidity (Diaz et al., 2018), stages 
of cancer (Acharya and Grigsby, 2016; Landy et al., 
2016), insurance (Acharya and Grigsby, 2016; 
Churilla et al., 2016 ; Davis et al., 2018; Fedewa et 
al., 2012) and distance (Barrington et al., 2016) 
have been reported as risk factors responsible for 
the incidence and mortality of cervical cancer. 

Identifying risk factors for cervical cancer is critical 
for developing proper prevention and treatment 
approaches connecting to risk factors, thus 
reducing racial disparities on the outcomes of 
cervical cancer, such as higher incidence and 
mortality rates, as well as lower five-year survival 
and overall survival among black women 
(American Cancer Society, 2018). 

There were significant differences in the 
characteristics between white and black women 

with cervical cancer in the NCDB data. For 
example, black women were diagnosed with later-
stage disease with more comorbidities than white 
women, which resulted in increased morbidity and 
mortality in black women. In addition, most black 
women had treatment in academic/research 
programs that were close to their residence and 
higher mortality rates among white women were 
identified in community cancer programs and 
Comprehensive Community Cancer Programs. 

Considering significant racial disparities in 
healthcare outcomes, it is essential to understand 
the impact of the type of health care facility on 
health outcomes among cervical cancer patients. 
Several studies have been conducted on the 
relationship between outcomes in cervical cancer 
patients based on distance from hospitals 
(Barrington et al., 2016; Gunderson et al., 2013; 
Powell et al., 2018; Spees et al., 2018) and the 
impact of facility volume on quality treatment and 
survival (Ross et al., 2010; Showalter et al., 2016). In 
their study on racial disparities in outcomes after 
surgical procedures, Haider and associates (2013) 
reported systemic factors such as access to care, 
hospital volume, and hospital patient population 
have been shown to contribute to disparities 
(Haider et al., 2013). Study results report that high 
volume hospitals have more favorable outcomes 
than low volume hospitals. 

Connecting to the choice of hospitals for 
treatment, it is also important to understand 
possible factors affecting the patient’s choice of 
treatment options resulting in the subsequent 
oncologic outcome and causing significant 
disparities in survival among minority patients (Luo 
et al., 2015). For example, black patients and living 
in rural areas were associated with the choice of 
surgery to treat other cancers, resulting in less 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e3                                              Cancer Health Disparities 

RESEARCH 

surgery compared to white patients (Cykert et al., 
2010; Steenland et al., 2011). 

The purpose of this study was to investigate the 
impact of different health care facilities on types of 
treatments and cervical cancer outcomes using the 
National Cancer Database (NCDB) data. The 
research question was “How would different types 
of health care facilities influence the outcomes of 
patients with cervical cancer and their types of 
treatment?” 

Materials and Methods 

To assess cervical cancer outcomes by different 
facility types, cervical cancer data were drawn from 
the National Cancer Database (NCDB) for the 
years 2004 to 2015. NCDB is jointly sponsored by 
the American College of Surgeons and the 
American Cancer Society, and it is the largest 
clinical cancer registry in the world (Steenland et 
al., 2011). It covers more than 70% of newly 
diagnosed cancer cases in the United States (Lin et 
al., 2014). 

Each facility reporting cases to the NCDB is 
assigned to a category classification, i.e., the facility 
type, by the Commission on Cancer Accreditation 
program. The NCDB classified all facilities into four 
types: Academic Cancer Programs, Community 
Cancer Programs, Comprehensive Community 
Cancer Programs, and Integrated Network Cancer 
Programs (Source: http://ncdbpuf.facs.org/content 
/participant-use-file-facility-type). Academic 
Comprehensive Cancer Program or 
Academic/Research programs (ARPs) participate in 
postgraduate medical education in at least four 
program areas, including internal medicine and 
general surgery. The facility accessions more than 
500 newly diagnosed cancer cases each year, and 
offers the full range of diagnostic and treatment 

services either on-site or by referral. Community 
Cancer Programs (CCPs) accession more than 100 
but fewer than 500 newly diagnosed cancer cases 
each year and provides a full range of diagnostic 
and treatment services, but referral for a portion of 
diagnosis or treatment may occur. Comprehensive 
Community Cancer Programs (CCCPs) accession 
500 or more newly diagnosed cancer cases each 
year and provide a full range of diagnostic and 
treatment services either on-site or by referral. 
Integrated Network Cancer Programs (INCPs) own, 
operate, lease, or are part of a joint venture with 
multiple facilities providing integrated cancer care 
and offer comprehensive services. We used this 
classification to analyze demographic 
characteristics, disease status, treatment and 
cervical cancer outcomes by the type of facility. 

For our data analysis, women younger than 40 
years old were excluded from this study in addition 
to women other than white and black (Hispanic 
origin was not distinguished). Age was categorized 
into 5 groups: 40-49, 50-59, 60-69, 70-79, and 
80+. Regional information was deduced from the 
location of the reporting facility (suppressed for 
cases aged 0-39). 

There are 6 categories for insurance in NCDB: 
uninsured, private insurance/managed care, 
Medicaid, Medicare, other government, and 
unknown. We further separated the Medicare 
group into younger Medicare (< 65 years old) and 
older Medicare (≥ 65 years old) because of the 
difference in eligibility for these two groups. 
Cervical cancer stages were identified based on 
the T, N, and M elements as defined by the 
American Joint Committee on Cancer (AJCC). To 
analyze the association between disease status 
and facility types, “TNM Clin Stage Group” that 
identifies the anatomic extent of disease based on 

http://ncdbpuf.facs.org/content/participant-use-file-facility-type
http://ncdbpuf.facs.org/content/participant-use-file-facility-type


 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e4                                              Cancer Health Disparities 

RESEARCH 

the T, N, and M elements known prior to the start 
of any therapy was re-categorized into 5 
groups:1A-1B1 (TNM Clin Stage Group=1,1A,1A1, 
1A2,1B,1B1), 1B2 (TNM Clin Stage Group=1B2), 2 
(TNM Clin Stage Group=2,2A,2A1,2A2,2B), 3 (TNM 
Clin Stage Group=3,3A,3B), and 4 (TNM Clin Stage 
Group=4,4A,4B). 

Comorbidity was measured by Charlson score, 
which is a weighted score derived from the sum of 
the scores for each of the comorbid conditions 
listed in the Charlson Comorbidity Score Mapping 
Table. 

Household income was categorized as quartiles 
based on equally proportioned income ranges 
among all US zip codes. Education was estimated 
by matching the zip code of the patient recorded 
at the time of diagnosis against files derived from 
the 2012 American Community Survey data, 
spanning years 2008-2012. This item provides a 
measure of the number of adults in the patient's 
zip code who did not graduate from high school, 
and is categorized as equally proportioned 
quartiles among all US zip codes. 

Residence, which represents the area-based 
measure of rurality and urban influence, was 
estimated by matching the state and county FIPS 
code of the patient recorded at the time of 
diagnosis against 2013 files published by the 
United States Department of Agriculture Economic 
Research Service. Metro counties and urban 
counties (as defined in the 2014 NCDB Participant 
Use Data File (PUF) Data Dictionary) were 
combined together as “Urban”, and compared 
with rural counties. 

SAS version 9.4, a software package for statistical 
analysis, was used to compute descriptive statistics. 
Chi-Square tests were used to compare the 

differences in demographic characteristics, disease 
status (cervical cancer stage at diagnosis), 
treatment, and cervical cancer outcomes among 
programs. Multivariable analysis was performed 
using stepwise modeling for variables associated 
with the outcomes of cervical cancer including vital 
status and 5 year survival. Odds ratio with 95% 
confidence intervals were calculated. 

Results 

Table 1 presents the association between 
demographic information and facility types. About 
forty-four percent of women used ARPs, followed 
by CCCPs (37.4%), INCPs (11.4%), and CCPs (6.8%). 

ARPs had more black women and younger women 
aged 40-59 years at diagnosis compared to CCPs 
and CCCPs, had less income and education (more 
women from the area who did not graduate from 
high school) compared to CCCPs and INCPs, more 
income compared to CCPs, more living far from 
hospitals, more uninsured, more lived in Midwest 
and Northeast compared to all other programs. 

CCPs had more women aged 75 and older when 
diagnosed compared to ARPs and INCPs, income 
with $38,000-47,999, education with 13-20.9% who 
did not graduate from high school, more living in 
rural areas and less than 10 miles away from the 
facility compared to all other programs. 

CCCPs had more white women compared to ARPs 
and INCPs, more women aged 65-74 years when 
diagnosed compared to ARPs, more education 
with less than 7% and with 7-12.9% who did not 
graduate from high school compared to ARPs and 
CCPs, more living in 21-50 miles away from the 
facility compared to all other programs, more 
private and other government insurance program 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e5                                              Cancer Health Disparities 

RESEARCH 

compared to ARPs and CCPs, and more living in 
West region compared to all other programs. 

INCPs had more income with $48,000-$62,999 
compared to ARPs and CCPs, and more living in 

urban, 11-20 miles from the facility and in South 
region compared to all other programs. 

Table 1. The demographic information by facility types (NCDB, 2004-2015). 
  Type of facility 

Demo 
Facility Type Total  

N (%) ARPs CCPs CCCPs INCPs 
Facility   35719(44.43) 5447(6.78) 30050(37.38) 9169(11.41) 80385 
Race* White  28096(78.66) 4670(85.74) 25911(86.23) 7328(79.92) 66005 

Black 7623(21.34) 777(14.26) 4139(13.77) 1841(20.08) 14380 
Age*† 40-44 6489(18.17) 914(16.78) 5088(16.93) 1636(17.84) 14127 

45-49 6466(18.10) 883(16.21) 4937(16.43) 1567(17.09) 13853 
50-54 5690(15.93) 818(15.02) 4416(14.70) 1406(15.33) 12330 
55-59 4970(13.91) 699(12.83) 4016(13.36) 1262(13.76) 10947 
60-64 3956(11.08) 609(11.18) 3257(10.84) 1026(11.19) 8848 
65-69 2912(8.15) 472(8.67) 2779(9.25) 782(8.53) 6945 
70-74 2026(5.67) 354(6.50) 1987(6.61) 551(6.01) 4918 
75+ 3210(8.99) 698(12.81) 3570(11.88) 939(10.24) 8417 

Income* <$38,000 9807(27.46) 1452(26.66) 6756(22.48) 2172(23.69) 20187 
$38,000-47,999 8356(23.39) 1669(30.64) 7961(26.49) 2245(24.48) 20231 
$48,000-62,999 8450(23.66) 1322(24.27) 7841(26.09) 2444(26.66) 20057 
>=$63,000 8733(24.45) 942(17.29) 7095(23.61) 2183(23.81) 18953 
Unknown 373(1.04) 62(1.14) 397(1.32) 125(1.36) 957 

Education*†† <7% 5646(15.81) 606(11.13) 5236(17.42) 1561(17.02) 13049 
7-12.9% 9094(25.46) 1530(28.09) 9099(30.28) 2752(30.01) 22475 
13-20.9% 9928(27.79) 1709(31.38) 8763(29.16) 2761(30.11) 23161 
>=21% 10694(29.94) 1542(28.31) 6574(21.88) 1971(21.50) 20781 
Missing 357(1.00) 60(1.10) 378(1.26) 124(1.35) 919 

Residence* Urban 34135(95.57) 5125(94.09) 28463(94.72) 8819(96.18) 76542 
Rural 489(1.37) 195(3.58) 656(2.18) 61(0.67) 1401 
Missing 1095(3.07) 127(2.33) 931(3.10) 289(3.15) 2442 

Distance* 
 

<=10 15743(45.65) 3301(62.80) 14379(49.80) 4757(53.80) 38180 
11-20 5578(16.18) 981(18.66) 5250(18.18) 1702(19.25) 13511 
21-50 6742(19.55) 742(14.12) 5758(19.94) 1585(17.93) 14827 
>50 6421(18.62) 232(4.41) 3486(12.07) 798(9.03) 10937 

Insurance* Uninsured 3829(10.78) 424(7.83) 2113(7.06) 800(8.77) 7166 
Medicaid 7373(20.76) 1273(23.50) 4320(14.44) 1601(17.55) 14567 
Younger 
Medicare 

1579(4.45) 312(5.76) 1343(4.49) 368(4.03) 3602 

Older Medicare 6039(17.00) 1265(23.35) 6815(22.79) 1863(20.42) 15982 
Private 14866(41.85) 1979(36.53) 14443(48.29) 4262(46.71) 35550 
Other 301(0.85) 41(0.76) 371(1.24) 109(1.19) 822 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e6                                              Cancer Health Disparities 

RESEARCH 

Government 
Missing 1536(4.32) 123(2.27) 504(1.69) 121(1.33) 2284 

Region*  Midwest 9076(25.41) 1777(32.62) 6209(20.66) 1867(20.36) 18929 
Northeast 9083(25.43) 900(16.52) 4534(15.09) 873(9.52) 15390 
South 12833(35.93) 1928(35.40) 13429(44.69) 5421(59.12) 33611 
West 4727(13.23) 842(15.46) 5878(19.56) 1008(10.99) 12455 

* P<.0001 
† All women whose age were less than 40 years were missing in this facility analysis (26,927, 25.1%). 
†† Education was measured using the number of adults in the patient's zip code who did not graduate from high 
school and is categorized as equally proportioned quartiles among all US zip codes. 
 
TNM Clin Stage Group presented that ARPs had more women in Stage 2, CCPs in Stage 4, and INCPs in 
Stage 1B2 compared to all other programs. ARPs had more women with no comorbidity while INCPs had 
more women with 1 comorbidity compared to all other programs (Table 2).  

Table 2. Disease status by facility types (NCDB, 2004-2015). 
  Type of facility 
Disease status 

Facility Type Total  
N (%) ARPs CCPs CCCPs INCPs 

TNM Clin Stage 
Group* 

1A-1B1 9357(32.92) 1293(30.71) 7702(33.67) 2324(33.71) 20676 
1B2 1516(5.33) 176(4.18) 1186(5.18) 405(5.87) 3283 
2 5806(20.42) 784(18.62) 4270(18.67) 1313(19.04) 12173 
3 6885(24.22) 989(23.49) 5456(23.85) 1692(24.54) 15022 
4 4862(17.10) 968(22.99) 4262(18.63) 1161(16.84) 11253 

Comorbidity* 0 29874(84.61) 4507(83.66) 24861(83.59) 7434(82.19) 66676 
1 4521(12.81) 722(13.40) 4031(13.55) 1361(15.05) 10635 
2 911(2.58) 158(2.93) 849(2.85) 250(2.76) 2168 

* P<.0001 

The association between treatment types and race 
by the types of facility was analyzed. Black women 
had more surgery, radiation, and chemotherapy in 
ARPs compared to all other programs, while white 
women had more surgery and chemo in CCCPs 
compared to ARPs and INCPs and more radiation 
in CCPs compared to ARPs and INCPs.  

Women who did not have surgery were about five 
times as likely to die of cervical cancer (Odds ratio 

4.90, 95%CI: 4.74-5.06), while women who did not 
have radiation or chemotherapy were less likely to 
die of cervical cancer (Odds ratio 0.84, 95%CI: 
0.81-0.87; Odd ratio 0.86, 95%CI:0.82-0.89) (data 
not shown). 

INCPs had more women with surgery compared 
to all other programs. 

 
 
 
Table 3. Treatment status by race and facility types (NCDB, 2004-2015). 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e7                                              Cancer Health Disparities 

RESEARCH 

  Type of facility 
Treatment 

Facility Type Total  
N (%) ARPs CCPs CCCPs INCPs 

Surgery* Yes  16432(46.10) 2293(42.40) 14818(49.51) 4682(51.22) 
      White  13597(82.75) 2013(87.79) 13126(88.58) 3910(83.51) 32646 
      Black  2835(17.25) 280(12.21) 1692(11.42) 772(16.49) 5579 
No 19210(53.90) 3115(57.60) 15113(50.49) 4459(48.78) 41897 

Chemo*  Yes 22110(62.43) 3328(62.03) 18034(60.73) 5641(62.13) 49113 
      White 17165(77.63) 2836(85.22) 15487(85.88) 4479(79.40) 39967 
      Black 4945(22.37) 492(14.78) 2547(14.12) 1162(20.60) 9146 
No 13308(37.57) 2037(37.97) 11663(39.27) 3439(37.87) 30447 

Radiation* Yes 22470(63.31) 3425(63.50) 18547(62.26) 5612(61.51) 50054 
      White 17409(77.48) 2940(85.84) 15832(85.36) 4453(79.35) 40634 
      Black 5061(22.52) 485(14.16) 2715(14.64) 1159(20.65) 9420 
No 13024(36.69) 1969(36.50) 11242(37.74) 3511(38.49) 29746 

* P<.0001 

Five outcomes were analyzed to identify their 
association with facility types. ARPs had more women 
alive at 30 and 90 days after surgery compared to 
CCPs, while INCPs had more cases died less than 30 
days after surgery compared to ARPs and CCCPs. CCPS 
had more cases that died fewer than 90 days after 
surgery, more cases with no surgery or alive cases that 
had fewer than 30 and 90 days of follow up and more 
deceased cases compared to all other programs. 

CCCPs had more cases with no surgery or alive cases 
that had fewer than 30 days of follow up compared to 
all other programs. 

Regarding the outcome of readmission within 30 days 
of surgical discharge, overall 94.6% was either no 

surgical procedure of the primary site was performed, 
or patient was not readmitted. 

INCPs had more cases with unplanned readmission 
within 30 days of discharge compared to all other 
programs.  

ARPs presented more five-year survival compared to 
CCPs and less comorbidity compared to INCPs, while 
CCPs had less five year survival compared to all other 
programs. Regarding the outcome of vital status 
(mortality), ARPs had more cases who were alive 
compared to CCPs and CCCPs, while CCPs had more 
cases who were deceased compared to all other 
programs (Table 4). 

 

Table 4. Survival outcomes by facility types (NCDB, 2004-2015). 
  Type of facility 

Outcomes 
Facility Type Total  

N (%) ARPs CCPs CCCPs INCPs 
30 Day 
Mortality* 

Alive at 30 days after surgery 
Died < 30 days after surgery 
No surgery, or alive cases 
have < 30 days of follow up 

14681(98.38) 1988(96.18) 13102(97.81) 4127(98.10) 33898 
56(0.38) 11(0.53) 56(0.42) 25(0.59) 148 
186(1.25) 68(3.29) 237(1.77) 55(1.31) 546 

90 Day 
Mortality* 

Alive at 90 days after surgery 
Died < 90 days after surgery 
No surgery, or alive cases 
have < 90 days of follow up 

14461(96.90) 1949(94.29) 12902(96.32) 4052(96.32) 33364  
168(1.13) 34(1.64) 173(1.29) 62(1.47) 437 

294(1.97) 84(4.06) 320(2.39) 93(2.21) 791 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e8                                              Cancer Health Disparities 

RESEARCH 

Readmission 
Within 30 
Days of 
Surgical 
Discharge* 

No surgical procedure 
Unplanned readmission  
Planned readmission  
Planned and unplanned 
readmission  
Unknown 

33713(94.38) 5111(93.83) 28519(94.91) 8675(94.61) 76018 
690(1.93) 87(1.60) 454(1.51) 190(2.07) 1421 
417(1.17) 66(1.21) 278(0.93) 102 (1.11) 863 
33(0.09) 2(0.04) 23 (0.08) 5(0.05) 63 

866(2.42) 181(3.32) 776(2.58) 197(2.15) 2020 

5 Year 
survival* 

Yes 
No 

10895(33.27) 1471(29.53) 8918(32.49) 2684(32.28) 23968 
21855(66.73) 3511(70.47) 18531(67.51) 5630(67.72) 49527 

Vital status 
(Mortality) * 

Dead 
Alive 

12456(38.03) 2225(44.63) 11480(41.82) 3219(38.72) 29380 
20299(61.97) 2760(55.37) 15971(58.18) 5095(61.28) 44125 

* P<.0001 

Table 5 presents the result of multivariate logistic 
regression on vital status (mortality) and 5-year survival 
in cervical cancer. There were slight differences in 
predicting vital status and 5-year survival. 

Black women, insurances other than private insurance, 
less income, more education, more comorbidity, more 
advanced stages, older women, treatment in CCPS and 
CCCPs, and not having surgery and chemo were more 
likely to die of cervical cancer compared to the 
reference groups. 

Black women, insurances other than private insurance, 
more education, more comorbidity, more advanced 
stages, living far from facilities, older women, treatment 
in CCCPs and INCPs, and not having surgery and 
radiation were less likely to survive for five years 
compared to the reference groups. 

Although distance in predicting vital status and income 
in predicting 5-year mortality were not significant from 
stepwise, we added them to the model to adjust 
confounding effects. 

Table 5. Predictors of the outcomes of cervical cancer (NCDB, 2004-2015). 
 Vital status (Mortality) 5 year survival 
 Adjusted 

Odds Ratio 
95% 

confidence 
interval 

Adjusted 
Odds Ratio 

95% 
confidence 

interval 
Race  
      White 
      Black  

 
1.00  
1.10 

 
- 
1.04-1.16 

 
1.00 
0.94 

 
- 
0.89-0.99 

Region  
      South 
      Midwest 

 
1.00 
1.03 

 
- 
0.98-1.09 

 
1.00 
1.03 

 
- 
0.97-1.08 

      Northeast  0.94 0.88-1.00 1.01 0.95-1.07 
      West  0.94 0.89-1.01 1.10 1.04-1.17 
Insurance  
      Private 
      Uninsured  

 
1.00 
1.03 

 
- 
0.96-1.11 

 
1.00 
0.80 

 
- 
0.74-0.86 

      Medicaid 1.24 1.17-1.31 0.75 0.71-0.79 
      Younger Medicare  1.49 1.35-1.64 0.70 0.63-0.78 
      Older Medicare  1.54 1.40-1.70 0.84 0.75-0.93 
      Other Government  0.85 0.69-1.05 0.85 0.70-1.03 
      Missing  1.05 0.94-1.19 0.92 0.82-1.03 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e9                                              Cancer Health Disparities 

RESEARCH 

Income  
      <$38,000 
      $38,000-47,999  

 
1.00 
0.90 

 
- 
0.84-0.95 

 
1.00 
1.03 

 
- 
0.97-1.09 

      $48,000-62,999  0.86 0.80-0.92 1.03 0.96-1.10 
      >=$63,000  0.78 0.72-0.84 1.05 0.96-1.13 
      Unknown  1.02 0.40-2.58 1.03 0.41-2.62 
Education†  
      >=21% 
      <7% 

 
1.00 
1.12 

 
- 
1.03-1.23 

 
1.00 
0.94 

 
- 
0.86-1.02 

      7-12.9%  1.16 1.08-1.24 0.93 0.87-0.99 
      13-20.9% 1.13 1.07-1.20 0.98 0.92-1.03 
      Missing  2.60 1.01-6.68 0.30 0.11-0.78 
Comorbidity 
      0 
      1  

 
1.00 
1.37 

 
- 
1.29-1.45 

 
1.00 
0.79 

 
- 
0.74-0.84 

      2  1.98 1.75-2.24 0.60 0.52-0.69 
Stage 
      1A-1B1 
      1B2 

 
1.00 
1.34 

 
- 
1.21-1.49 

 
1.00 
0.64 

 
- 
0.59-0.71 

      2  1.84 1.72-1.97 0.76 0.71-0.81 
      3  3.62 3.38-3.87 0.42 0.39-0.45 
      4  11.52 10.68-12.43 0.12 0.10-0.13 
Distance(miles)     
      <=10  
      11-20  

1.00 
0.99 

- 
0.94-1.05 

1.00 
0.94 

- 
0.89-0.99 

      21-50  1.00 0.95-1.06 0.91 0.86-0.96 
      >50  1.05 0.98-1.11 0.82 0.77-0.87 
Age 
      40-44 
      45-49  

 
1.00 
1.20 

 
- 
1.12-1.29 

 
1.00 
0.99 

 
- 
0.93-1.06 

      50-54  1.27 1.18-1.36 0.95 0.89-1.02 
      55-59  1.38 1.28-1.49 0.88 0.82-0.95 
      60-64  1.52 1.41-1.65 0.85 0.79-0.92 
      65-69  1.51 1.36-1.69 0.84 0.75-0.94 
      70-74  1.91 1.69-2.15 0.71 0.63-0.81 
      75+  3.53 3.15-3.96 0.46 0.40-0.51 
Facility  
     Academic/Research Program 
     Community Cancer Program  

 
1.00 
1.16 

 
- 
1.07-1.27 

 
1.00 
0.93 

 
- 
0.86-1.02 

     Comprehensive Community Cancer Program  1.19 1.14-1.25 0.94 0.90-0.99 
     Integrated Network Cancer Program  1.04 0.97-1.11 0.87 0.81-0.93 
Surgery 
     Yes 
     No  

 
1.00 
1.90 

 
- 
1.80-2.01 

 
1.00 
0.77 

 
- 
0.73-0.82 

Radiation 
     Yes 

 
1.00 

 
- 

 
1.00 

 
- 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e10                                              Cancer Health Disparities 

RESEARCH 

     No 0.90 0.85-0.96 0.70 0.66-0.75 
Chemo 
     Yes 
     No  

 
1.00 
1.16 

 
- 
1.09-1.22 

 
1.00 
1.09 

 
- 
1.03-1.15 

† Education was measured using the number of adults in the patient's zip code who did not graduate from high 
school and is categorized as equally proportioned quartiles among all US zip codes. 

Discussion 

To prevent and treat cervical cancer effectively, it is 
critical to understand risk factors and how they 
impact the disease. While assessing the differences 
in risk factors and outcomes of cervical cancer 
between races in our previous study, we found 
there were differences in using health care facilities 
between black and white women. Therefore, we 
further analyzed the NCDB data to identify 
differences in demographics, disease status, 
treatment status and health outcomes by facility 
type. 

ARPs had more black women, women younger at 
diagnosis with less education, more living far from 
treatment facilities, more in Stage 2, less 
comorbidity and more 5-year survival, and more 
cases alive at 30 and 90 days after surgery 
compared to other programs. CCPs had more 
women aged 75 and older at diagnosis, living in 
rural areas and fewer than 10 miles from hospitals, 
more Stage 4 and comorbidity, less 5 year survival, 
and more radiation compared to other programs. 
CCCPs had more white women, more education 
and more private insurance, and more surgery. 
INCPs had more women living in urban and south 
region, more Stage 1B2 and more surgery, more 1 
comorbidity, and more women who died less than 
30 days after surgery. 

Previous studies reported having treatment at 
high-volume hospitals was associated with lower 
mortality rates and better survival (Haider et al., 

2013; Lin et al., 2014). However, when assessing the 
impact of the type and volume of facility on health 
care outcome, it is critical to consider patients’ 
disease status. ARPs that would be classified as 
high volume centers presented better outcomes in 
being alive at 30 and 90 days after surgery, 5-year 
survival and, vital status, compared to CCPs that 
would be classified as low volume centers. 
However, it was noted that CCPs had more 
women diagnosed at Stage 4 while ARPs had 
more women diagnosed at Stage 2, so the result 
should be interpreted carefully. 

It is also interesting that younger age groups had 
more treatment in ARPs while older age groups 
had more treatment in CCPs and CCCPs. That may 
explain the possible association with a better 
survival outcome in cervical cancer in ARPs, which 
is consistent with prior works (Fedewa et al., 2012; 
Furlow, 2018; Yosta and Hoekstra, 2018). 

From multivariate analysis results, under-insured 
women were more likely to die of cervical cancer 
and, uninsured and under-insured women were 
less likely to survive for five years compared to 
privately insured women. It is interesting that the 
uninsured women were not more likely to die of 
cervical cancer compared to private patients. 
Although women without health insurance are less 
likely to receive cervical cancer treatment and 
survive (Acharya and Grigsby, 2016; Fedewa et al., 
2012), our analysis shows over 53% of uninsured 
patients were treated in ARPs and had better 
health outcomes in 30 day and 90 day mortality 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e11                                              Cancer Health Disparities 

RESEARCH 

and vital status, which proves promising in 
developing cervical cancer centers that may 
improve healthcare access and provide a high 
quality of care to those with under- or no 
insurance. 

However, when the population was stratified by 
race within ARPs, black women had more deaths 
less than 30 days and 90 days after surgery 
(respectively p<.05) and had more deaths (Vital 
status: p<.0001) compared to white women. 
Further analyses showed black women in ARPs 
were older, less income, more living in urban area 
and close to treatment facilities, more uninsured 
and underinsured, more advanced stages (Stage 3 
& 4), more comorbidities, less surgery, and more 
chemo and radiation therapy compared to white 
women (data not shown). Considering the 
proportion of black women (21%) and still overall 
better outcomes for women with cervical cancer in 
ARPs, it addresses clear cervical cancer outcome 
disparities between black and white women 
(Arvizo and Mahdi, 2017; Dalton & Farley, 2017; 
Fleming and et al., 2014). 

Therefore, it is needed to assess the effectiveness 
of cervical cancer treatment/programs according 
to the facility type, especially, to explore how to 
manage patients without insurance and with less 
income and how to reduce outcome disparities by 
targeting barriers identified. 

As another interesting result, we found there was 
an interaction between education and income 
(Data not shown) that women from higher 
education areas had less survival, which is 
contradicting to previous study results (Franceschi 
et al., 2009; Singh and Jemal, 2017). This might be 
associated with the fact that education was 
measured using the number of adults in the 
patient's zip code who did not graduate from high 

school, not based on an individual’s educational 
attainment. 

Therefore, further study is necessary to investigate 
how the level of education and income are related 
to cervical cancer risk and mortality. 

Our study has some limitations especially in 
assessing the outcomes of cervical cancer in each 
facility. We compared the association between 
cervical cancer outcomes and facility type based 
on responses from a large national sample of 
cancer patients, so our results should not be 
interpreted as actual evaluation of cervical cancer 
treatment in each facility classified by NCDB. Also, 
considering a disproportionate number of women 
treated at each facility, for example, 44.4% patients 
at ARPs compared with about 6.8% at CCPs, the 
generalizability of this study may be limited. More 
importantly, a significant number of patients had 
missing data on facility at each stage of their 
diagnosis (e.g., 34.99% did not have facility 
information in Stage 1A-1B1). 

As another limitation, our analysis was limited by 
the variables collected in the NCDB. For example, 
there was no data to assess factors contributing to 
choosing a facility and treatment, so the reason for 
racial differences in using different facilities could 
not be explored. 

Considering a complex situation that patients with 
cervical cancer may confront, it is imperative to 
understand what factors contribute to choosing 
their treatment option including health care facility. 
For example, possible reasons resulting from 
limited access to high-volume/quality hospitals, 
such as geographic location and under- or no 
insurance among minorities leading to outcome 
disparities, should be considered and reflected 
when developing cervical cancer centers and 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e12                                              Cancer Health Disparities 

RESEARCH 

programs (Beavis et al., 2017; Haider et al., 2013; 
Singh and Jemal, 2017). 

Our study results include some contradicting 
outcomes, which may encourage further research 
on understanding treatment choices and 
outcomes better. This will help identify critical 
factors that make cancer care programs successful 
and effective. 

Acknowledgements 
This article was partially supported by the 
National Cancer Institute (P30CA013148, 
P50CA098252, 2U54CA118948). The contents of 
this article are solely the responsibility of the 
authors and do not necessarily represent the 
official views of the funding agency, NIH. 

Conflict of interest 
The authors declare that there are no conflicts of 
interest or financial disclosures. 

Authors’ contributions 
Conceptualization: HP SB CW 

Formal analysis: HP ZW SB 

Methodology: HP SB 

Supervision: SB WH 

Writing ± original draft: HP SB 

Writing ± review & editing: HP ZW CW WH SB 

REFERENCES 
Acharya, S and Grigsby, P.W., Access to Health Care and 

Disparities in Cervical Cancer Diagnosis, Treatment, and 
Survival. International Journal of Radiation Oncology, 
2016. 96(2): p. e290. 

Akinlotan M, B.J., Helduser J, Ojinnaka C, Lichorad A, and 
McClellan D, Cervical Cancer Screening Barriers and Risk 
Factor Knowledge Among Uninsured Women. J 
Community Health, 2017. 42(4): p. 770–778. 

American Cancer Society, Cancer Facts & Figures for African 
Americans 2016-2018. Atlanta: American Cancer Society, 
2018. 

Arvizo, C., and Mahdi, H., Disparities in cervical cancer in 
African American women: What primary care physicians 
can do. Cleve Clin J Med, 2017. 84(10): p. 788-794. 

Barrington, D.A., Dilley, S.E., Landers, E.E., Thomas, E.D., 
Boone, J.D., Straughn, J.M. Jr, McGwin, G. Jr, and Leath, 
C.A., Distance from a Comprehensive Cancer Center: A 
proxy for poor cervical cancer outcomes? Gynecol 
Oncol., 2016. 143(3): p. 617-621. 

Beavis, A.L., Gravitt, P.E. and Rositch, A.F., Hysterectomy-
corrected cervical cancer mortality rates reveal a larger 
racial disparity in the United States. . Cancer, 2017. 123(6): 
p. 1044-1050. 

CDC, HPV Vaccine. Accessed on Dec 20, 2018 
https://www.cdc.gov/cancer/knowledge/provider-
education/hpv/hpv-vaccine.htm, 2018. 

Churilla, T., Egleston, B., Dong, Y., Shaikh, T., Murphy, C., 
Mantia-Smaldone, G., Chu, C., Rubin, S., and Anderson, 
P., Disparities in the management and outcome of 
cervical cancer in the United States according to health 
insurance status. Gynecol Oncol, 2016 141(3): p. 516-523. 

Cykert, S., Dilworth-Anderson, P., Monroe, M.H., Walker, P., 
McGuire, F.R., Corbie-Smith, G., Edwards, L.J., and 
Bunton, A.J., Factors associated with decisions to 
undergo surgery among patients with newly diagnosed 
early-stage lung cancer. JAMA, 2010 303(23): p. 2368-76. 

Dalton, H.J., and Farley, J.H., Racial disparities in cervical 
cancer: Worse than we thought. Cancer, 2017. 123(6): p. 
915-916. 

Davis, M., Strickland, K., Easter, S.R., Worley, M. Jr, Feltmate, 
C., Muto, M., Horowitz, N., Berkowitz, R., and Feldman, 
S., The impact of health insurance status on the stage of 
cervical cancer diagnosis at a tertiary care center in 
Massachusetts. Gynecol Oncol., 2018. 150(1): p. 67-72. 

Diaz, A., Baade, P.D., Valery, P.C., Whop, L.J., Moore, S.P., 
Cunningham, J., Garvey, G., Brotherton, J.M.L., O'Connell, 
D.L., Canfell, K., Sarfati, D., Roder, D., Buckley, E., and 
Condon, J.R., Comorbidity and cervical cancer survival of 
Indigenous and non-Indigenous Australian women: A 
semi-national registry-based cohort study (2003-2012). 
PLoS One, 2018. 13(5). 

Fedewa, S.A., Cokkinides, V., Virgo, K.S., Bandi, P., Saslow, D., 
and Ward, E.M., Association of Insurance Status and Age 
With Cervical Cancer Stage at Diagnosis. AJPH, 2012. 

Ferlay, J., Soerjomataram, I., Dikshit, R., Eser, S., Mathers, C., 
Rebelo, M., and Bray, F., Cancer incidence and mortality 
worldwide: Sources, methods and major patterns in 
GLOBOCAN 2012. International Journal of Cancer, 2015. 
136. 

Fleming, S., Schluterman, N.H., Tracy, J.K., and Temkin, S.M., 
Black and white women in Maryland receive different 
treatment for cervical cancer. PLoS One. 9(8): p. e104344. 



 
 
 
 
 

 
www.companyofscientists.com/index.php/chd                   e13                                              Cancer Health Disparities 

RESEARCH 

Franceschi, S., Plummer, M., Clifford, G., De Sanjose, S., Bosch, 
X., Herrero, R., Muñoz, N. and Vaccarella, S., Differences 
in the risk of cervical cancer and human papillomavirus 
infection by education level. British Journal of Cancer, 
2009. 101(5): p. 865. 

Furlow, B., Nearly One-in-Five Women With Cervical Cancer 
are Diagnosed After Age 65. 2018. 

Gunderson, C.C., N.E., McMeekin DS, and Moore KN., 
Distance traveled for treatment of cervical cancer: who 
travels the farthest, and does it impact outcome? Int J 
Gynecol Cancer, 2013 

Haider, A.H., Scott, V.K., Rehman, K.A., Velopulos, C., Bentley, 
J.M., Cornwell, E.E. and Al-Refaie, W., Racial disparities in 
surgical care and outcomes in the United States: a 
comprehensive review of patient, provider, and systemic 
factors. J Am Coll Surg, 2013. 216(3): p. 482-92. 

Landy, R., Pesola, F., Castañón, A., and Sasieni P., Impact of 
cervical screening on cervical cancer mortality: estimation 
using stage-specific results from a nested case–control 
study. Br J Cancer, 2016. 115(9): p. 1140-1146. 

Lin. J.F., B., J.L., Krivak, T.C., Beriwal, S., Chan, J.K., 
Sukumvanich, P., Monk, B.J., and Richard, S.D., Impact of 
facility volume on therapy and survival for locally 
advanced cervical cancer. Gynecol Oncol, 2014 132(2): p. 
416-22. 

Luo, T., Spolverato, G., Johnston, F., Haider, A.H., and Pawlik, 
T.M., Factors That Determine Cancer Treatment Choice 
Among Minority Groups. J Oncol Pract, 2015 11(3): p. 
259-61. 

Powell, T.C., Dilley, S.E., Bae, S., Straughn, J.M.Jr,, Kim, K.H., 
and Leath, C.A., The Impact of Racial, Geographic, and 
Socioeconomic Risk Factors on the Development of 
Advanced-Stage Cervical Cancer. J Low Genit Tract Dis, 
2018 22(4): p. 269-273. 

Ross, J.S., Normand, S.L., Wang, Y., Ko, D.T., Chen, J., Drye, 
E.E., Keenan, P.S., Lichtman, J.H., Bueno, H., Schreiner, 
G.C., and Krumholz, H.M., Hospital Volume and 30-Day 
Mortality for Three Common Medical Conditions. N Engl 
J Med, 2010. 362(12): p. 1110-8. 

Scarinci IC, G.F., Kobetz E, Partridge EE, Brandt HM, Bell MC, 
Dignan M, Ma GX, Daye JL, and Castle PE., Cervical 
cancer prevention: new tools and old barriers. Cancer, 
2010. . 116(11): p. 2531-42. 

Showalter, T.N., Camacho, F., Cantrell, L.A., and Anderson, 
R.T., Determinants of Quality Care and Mortality for 
Patients With Locally Advanced Cervical Cancer in 
Virginia. Medicine, 2016. 95(8): p. e2913. 

Singh, G.K., and Jemal, A., Socioeconomic and Racial/Ethnic 
Disparities in Cancer Mortality, Incidence, and Survival in 
the United States, 1950-2014: Over Six Decades of 
Changing Patterns and Widening Inequalities. J Environ 
Public Health, 2017. 2017. 

Spees, L.P., Wheeler, S.B., Varia, M., Weinberger, M., Baggett, 
C.D., Zhou, X., Petermann, V.M., and Brewster, W.R., 
Evaluating the urban-rural paradox: The complicated 
relationship between distance and the receipt of 
guideline-concordant care among cervical cancer 
patients. Gynecol Oncol, 2018 12(18): p. 31407-0. 

Steenland, K., Goodman, M., Liff, J., Diiorio, C., Butler, S., 
Roberts, P., Smith, J.L., Ekwueme, D. and Hall, I.J., The 
effect of race and rural residence on prostate cancer 
treatment choice among men in Georgia. Urology, 2011 
77(3): p. 581-7. 

Weragoda, J., Azuero, A., Badiga, S., Bell, W.C., Matthews, R., 
and Piyathilake, C., An examination of racial differences in 
5-year survival of cervical cancer among African 
American and white American women in the 
southeastern US from 1985 to 2010. Cancer Med, 2016 
5(8): p. 2126-35. 

Yoo, W., et al., Recent trends in racial and regional disparities 
in cervical cancer incidence and mortality in United 
States. PLoS One, 2017. 12(2): p. e0172548. 

Yosta and Hoekstra, A., Cervical cancer in women over 65: An 
analysis of screening. Gynecol Oncol Rep, 2018. 25: p. 
48–51. 

 
 

 
 
 


