Non-utilization of the Pap Test Among Women with Frequent Health System Contact

Abstract

Despite improvements in health access, many underserved women abstain from cervical cancer screening. A self-administered questionnaire was used to identify factors determining whether medically underserved women attending a safety net health system regularly are screened for cervical cancer. Approximately 11 % of study subjects had never received a Pap test despite an average of nearly four clinic visits in the preceding 12 months. Never screeners were significantly younger, more likely to be Hispanic, non-U.S. born and less likely to have healthcare continuity. In multivariable analysis, odds for never screening were independently lower among women with male partner support (aOR 0.29) and physician's recommendation for screening (aOR 0.34) and higher among women who believed screening visits are too long (aOR 2.53). Educating male partners of Hispanic and immigrant women in addition to addressing recognized situational barriers may help to improve cervical cancer screening rates.

Full text

ORIGINAL PAPER
Non-utilization of the Pap Test Among Women with Frequent
Health System Contact
Abayomi N. Ogunwale
1
•Haleh Sangi-Haghpeykar
1
•Jane Montealegre
2
•
Yiwen Cui
1
•Maria Jibaja-Weiss
3
•Matthew L. Anderson
4
Published online: 30 September 2015
ÓSpringer Science+Business Media New York 2015
Abstract Despite improvements in health access, many
underserved women abstain from cervical cancer screen-
ing. A self-administered questionnaire was used to identify
factors determining whether medically underserved women
attending a safety net health system regularly are screened
for cervical cancer. Approximately 11 % of study subjects
had never received a Pap test despite an average of nearly
four clinic visits in the preceding 12 months. Never
screeners were significantly younger, more likely to be
Hispanic, non-U.S. born and less likely to have healthcare
continuity. In multivariable analysis, odds for never
screening were independently lower among women with
male partner support (aOR 0.29) and physician’s recom-
mendation for screening (aOR 0.34) and higher among
women who believed screening visits are too long (aOR
2.53). Educating male partners of Hispanic and immigrant
women in addition to addressing recognized situational
barriers may help to improve cervical cancer screening
rates.
Keywords Cervical cancer Pap test Non-utilization 
Never-screened Male-partner support
Background
Population-based screening strategies have dramatically
reduced the burden of invasive cervical cancer in the
United States (U.S.). Despite the proven efficacy of these
strategies, however, many women remain unscreened. At
least one recent study has shown that the proportion of
women aged 22–30 in the U.S. who have never been
screened for cervical cancer increased from 6.6 to 9 %
between 2000 and 2010 [1]. Reasons for this increase
remain unclear. Nonetheless, measures to address this key
health disparity are important, as approximately 29–50 %
of all invasive cervical cancer diagnoses are made in
women who have never been previously screened with a
Pap test [2,3].
Financial barriers to health access have been repeatedly
shown to play an important role in determining rates of
cervical cancer screening. Recent mandates, such as the
Affordable Care Act (ACA) are widely anticipated to
improve rates of cervical cancer screening by requiring no-
cost access to preventive health services. Simulated anal-
yses of multi-state data from the Behavioral Risk Factor
&Matthew L. Anderson
matthew@bcm.edu
Abayomi N. Ogunwale
abayomi.ogunwale@bcm.edu
Haleh Sangi-Haghpeykar
hsaleh@bcm.edu
Jane Montealegre
jrmontea@bcm.edu
Yiwen Cui
yiwenc@bcm.edu
Maria Jibaja-Weiss
mariaj@bcm.edu
1
Department of Obstetrics and Gynecology, Baylor College of
Medicine, One Baylor Plaza, BCM611, Houston, TX 77030,
USA
2
Department of Pediatrics-Oncology, Baylor College of
Medicine, One Baylor Plaza, Houston, TX 77030, USA
3
Dan L. Duncan Cancer Center, Baylor College of Medicine,
One Baylor Plaza, Houston, TX 77030, USA
4
Department of Obstetrics and Gynecology, Pathology and
Immunology, Dan L. Duncan Cancer Center, Center for
Reproductive Biology, Baylor College of Medicine, One
Baylor Plaza, BCM 611, Houston, TX 77030, USA
123
J Immigrant Minority Health (2016) 18:1404–1412
DOI 10.1007/s10903-015-0287-9

Surveillance System (BRFSS) have predicted a 16 %
increase in Pap use among medically undeserved women
who have never been previously screened [4]. However,
there are many reasons in addition to a lack of health care
access why women may specifically choose not to partic-
ipate in cervical cancer screening. For example, screening
apathy appears to be increasing among U.S. women [2,5,
6]. Other personal and socio-cultural factors may also play
a role. These include age ([65), ethnicity (Hispanic and
other ethnic minorities), language barriers, low income,
and difficulties with communication access (e.g. lack of a
phone). In addition, ‘never-screened’ women are also more
likely to demonstrate poor knowledge of the Pap test and
risk factors for cervical cancer, and to hold negative or
incorrect beliefs about screening [7–10].
Recent studies have suggested that points of contact with
the health care system provide ideal opportunities to engage
women who have never been previously screened for cer-
vical cancer [5,11–13]. In one U.S. study, 90 % of women
who self-identified as having never received the Pap test
reported multiple clinical contacts with health providers in
the 5 years preceding survey [5]. A similar study in the
United Kingdom reported at least one clinic attendance per
year in[50 % of non-screeners [13]. Both of these studies
concluded that routine points of contact initiated for other
purposes are ideal venues for identifying and screening
women who have never previously had a Pap test. The
reasons why women with frequent health system contact fail
to participate in screening remain poorly understood. Thus,
it is not clear how best to engage this unique population of
at-risk women or whether concerted efforts to target never
screen women will be successful, particularly among
minority and immigrant populations.
Several small qualitative studies have suggested that
male partner opinions may be a barrier to cervical cancer
screening, especially among Hispanics [17–19]. However,
this observation has not been quantitatively validated. The
primary objective of this study was to assess male-partner
communication influences participation in cervical cancer
screening among women who have never been previously
screened. A secondary goal was to better understand other
psychosocial factors driving cervical screening non-par-
ticipation among women with frequent health system
contact.
Theoretical Framework
The theory of social support as proposed by Cassel and
Caplan [14,15] provided the conceptual basis for this
work. Social support has been shown to impact health
behaviours including breast and cervical cancer screening
among women [16].
Methods
Study Participants
We conducted a cross-sectional survey of consenting
women aged 18–64, presenting for care at two Harris Health
System (HHS) clinics with traditionally low rates of cer-
vical cancer screening. Harris Health is a safety-net health
system for indigent, medically underserved residents of
Harris County, Texas, a metropolitan statistical area with an
estimated population of 4.3 million people [20]. Harris
Health provides heavily subsidized care on a sliding scale
basis to Harris County residents having yearly incomes up
to 300 % above the Federal Poverty Limit (FPL) [21,22].
Data Collection
All women who checked-in at the two Harris Health
community health centers between November, 2013 and
January, 2014 were approached, screened for eligibility
and invited to participate in the study. Clinic sites were
selected due to their low rates of cervical cancer screening
and large numbers of patients seen each year ([10,000
distinct patient visits annually). Women who could not be
approached prior to their doctors’ appointments were
approached afterwards. Participants were given a three
paged questionnaire prepared in English or Spanish. Lan-
guage choice was left entirely at participants’ discretion. A
bilingual study coordinator fluent in both English and
Spanish was available to answer participants’ questions.
Measures
The survey instrument contained 39 items selected after an
extensive review of existing literature and previously pub-
lished surveys on the subject, including the BRFSS ques-
tionnaire [17,23]. Demographic items included age, race,
ethnicity, employment status, level of education and marital
status. Contact with the health system was assessed with
two questions: (1) ‘‘in the last 12 months how many times
have you visited a health care provider for a scheduled
appointment?’’ and (2) ‘‘are you usually seen by the same
doctor when you receive medical care?’’ ‘‘Yes/No’’. Par-
ticipants were considered to have frequent contact with the
health system if they reported [1 visit to an Harris Health
care provider in the preceding 12 months. Pap test receipt
was assessed by asking participants, ‘‘have you received a
Pap test in the (1) past 12 months, (2) past 3 years, (3) never
received?’’ For purposes of analysis, responses were
dichotomized as ‘‘never’’ (all women who reported never
screening) and ‘‘ever’’ (all other women who answered this
question, regardless of the timing of screening). Male
J Immigrant Minority Health (2016) 18:1404–1412 1405
123

partner influence on Pap utilization was assessed using two
different questions: (1) ‘‘does your partner want you to
receive regular Pap exam?’’ ‘‘Yes/No/I don’t know/He
doesn’t care’’, and (2) ‘‘I would be afraid to tell my partner
If I had cervical cancer, or received Pap smear, as it would
affect our relationship.’’ ‘‘Strongly agree/Somewhat agree/
Somewhat disagree/Strongly disagree.’’ Participants’
knowledge of cervical risk factors as well as beliefs, atti-
tudes and perceptions regarding Pap use were assessed
using items adapted from a validated study by Taylor et al.
and Byrd et al. [17,24]. Acculturation was assessed using
multiple commonly accepted measures such as self-re-
ported country of birth, duration of reported stay in the U.S.,
and the language in which participants chose to complete
the study questionnaire. Other items were included to assess
the influence of social support and situational barriers on
screening, as well as participants’ interactions with the
health system, continuity of care and existing medical
comorbidities. Corrections and additions were made to the
survey instrument following small focus groups interviews
of cervical cancer patients receiving care within the same
system, and reviews by local opinion leaders and Harris
Health health workers. Survey was initially developed in
English, translated to Spanish, and back-translated to Eng-
lish by native Spanish speakers to ensure lexical equiva-
lence. Bilingual women drawn from the local community
assessed the finished questionnaire to ensure appropriate
use of language and contexts.
This study was powered to assess the rate of Pap uti-
lization in the target population within 15 % of published
national rate (range 15–21 %) with 95 % confidence [25].
Eight hundred and thirty participants were targeted, and
sample size was increased to 1000, based on the
assumption that *20 % of respondents would fail to
respond to one or more key study questions. Approvals
for this study were provided by the Institutional Review
Boards of the Baylor College of Medicine (H-30980,
approved 9/24/2013) and Harris Health (13-11-0670,
approved 11/14/2013).
Analysis
Chi-square and Fisher exact tests were used to assess sta-
tistical significance for grouped variables, student’s ttest
for normally distributed continuous variables, and Wil-
coxon rank test for non-normal or ranked variables.
Logistic regression was used for estimating odds ratios
(OR) and 95 % confidence intervals for associations with
never having received a Pap test. A univariate analysis was
conducted initially. All potential confounders were asses-
sed and then controlled in a multivariable model. All
analyses were performed using SAS Statistical Software
(Version 9.3, SAS; Cary, NC). Associations with pB0.05
were considered statistically significant.
Results
Of the 1094 women invited to participate in this study,
1007 (92 %) consented. Twenty-four women did not
indicate the date of their most recent Pap test; responses
from these surveys were excluded from further analyses.
Approximately 11 % (n =106/983) of respondents repor-
ted that they had never received a Pap smear in their
lifetime. In univariate analyses, mean age of women who
reported never having had a Pap was significantly lower
than those with reported Pap screening experience (39.2 vs.
43.7 years; p=0.002). Never-testers were also more
likely to have been born outside the U.S. (76.9 vs. 58.7 %;
p=0.0003). Hispanics were proportionally over-repre-
sented among never screened women compared to the sub-
population of women with previous Pap screening experi-
ence (87.5 vs. 73.2 %; p=0.002). Pap utilization was not
shown to be associated to the total years of education
received, marital and employment status, as well as total
hours worked per week (Table 1). Never screened women
reported fewer hospital visits compared to screened women
(Table 2). Never screened women were also less likely to
have care continuity (i.e. seeing the same doctor on mul-
tiple clinic visits). Receipt of Pap test was not associated
with participants’ health insurance type, or use of mobility
equipment (Table 2). Women who reported never receiv-
ing the Pap test were less likely to report having a male
partner who wanted them to be screened (21.1 vs. 59.4 %;
p\0.0001; Table 3), and to report being afraid to tell their
partner of a cancer diagnosis or receipt of a Pap test
(Table 4). Similarly, never screened participants were less
likely than screened women to report having friends and
family members who receive the Pap test regularly (22 vs.
53.9 %; p=0.0001). They were also less likely to have
had a previous sexually transmitted infection (STI) or use
contraceptives compared to women with reported Pap
screening history (Table 3). No associations between
receipt of a Pap test and previous pregnancies, births, or
elective pregnancy terminations (Table 3).
Never screened women consistently demonstrated less
knowledge of the Pap smear, and cervical cancer risk
factors compared to screened women (Table 4). They were
also less likely to know that cervical cancer can be pre-
vented by doing regular Pap smears and that abnormal Pap
smears can be treated by a doctor. Knowledge of multiple
cervical cancer risk factors and beliefs were also associated
with whether or not study participants reported as never
screened (Table 4).
1406 J Immigrant Minority Health (2016) 18:1404–1412
123

In general, Pap screening behaviors among never screened
women were more influenced by situational barriers to
screening than among previously screened women. Never
screeners were more likely to consider Pap tests as too
expensive, to report communication problems with the health
provider, to consider screening visits too long and to report
having no one to care for their children as barriers to pap
screening compared to previously screened women (Table 4).
Table 1 Participants
demographics Pap never
Yes
N=106
No
N=877
p
Age, mean (SD) years 39.2 ±13.8 43.7 ±11.7 0.002
Education, mean (SD) years 10.8 ±3.8 10.9 ±3.4 0.89
Marital status n (% yes)
Married 51 (50) 469 (56.3) 0.62
Single, one partner 26 (25.5) 185 (22.2)
Single, [one partner 1 (0.9) 12 (1.4)
Single, no partner 24 (23.5) 167 (20.1)
Ethnicity, n (% yes) 0.004
Hispanic 91 (87.5) 624 (73.2) 0.002
African American 4 (3.9) 93 (10.9) 0.02
White 4 (3.9) 104 (12.2) 0.01
Asian/other 5 (4.8) 31 (3.6) 0.55
Birth country, n (% yes) 0.0003
USA 24 (23.1) 354 (41.3)
Non-USA 80 (76.9) 503 (58.7)
Employment
Employed, n (% yes) 31 (29.3) 299 (34.8) 0.26
Hours worked per week 17.9 ±16.9 22.4 ±19.2 0.11
Ability to request time-off, n (% yes) 35 (70) 382 (75.9) 0.35
Statistically significant pvalues are highlighted in bold
Unless otherwise specified, data are mean ±standard deviation or number (%) of women answering ‘‘yes’’
Table 2 Impact of health
access and continuity of care on
Pap smear use
Pap never
Yes
N=106
No
N=877
p
Health insurance type, n (% yes)
Medicare 3 (3.3) 41 (5.6) 0.38
Medicaid 5 (5.7) 53 (7) 0.63
Number of health visits in last 1 year, mean (SD) 3.4 ±2.9 4.1 ±3.4 0.04
Seen by the same physician (care continuity), n (% yes) 57 (57) 573 (67.1) 0.04
Pap smear recommended by physician, n (% yes) 20 (20.4) 493 (57.6) <0.0001
Use of durable mobility equipment, n (% yes) 12 (11.8) 119 (14.1) 0.52
Medical co-morbidities, n (% yes)
Diabetes 26 (25) 225 (25.6) 0.89
GI 13 (12.5) 69 (7.9) 0.10
Depression 10 (9.6) 157 (17.9) 0.03
Hypertension 17 (16.4) 243 (27.7) 0.01
Respiratory disease 3 (2.9) 43 (4.9) 0.46
Any cancer 1 (0.96) 48 (5.5) 0.05
Statistically significant pvalues are highlighted in bold
Unless otherwise specified, data are number and (%) of women answering ‘‘yes’’
J Immigrant Minority Health (2016) 18:1404–1412 1407
123

Logistic regression models adjusted for age, race and
U.S. birth showed similar results as the univariate analysis
(Table 5). To identify factors that independently predict
whether or not a woman has ever been screened for cer-
vical cancer, we conducted a final multivariable analysis in
which all significant variables in Table 5in addition to age,
ethnicity and birth country were included. Results of this
analysis identified three main variables common to never
screeners compared to previously screened women: lower
odds of reporting having partners who wanted them to have
regular Pap screening (aOR 0.29, 95 % CI 0.11–0.68;
p=0.006), lower odds of having a physician recommend a
Pap test (aOR 0.34, 95 % CI 0.14–0.80; p=0.02) and
higher odds of considering screening visits as too long
(aOR 2.53, 95 % CI 1.06–6.07; p=0.04; Table 6). No
other association remained significant in the final multi-
variable model.
Discussion
A key novel observation reported here is that perceptions
of male partner support play an important role in deter-
mining whether or not a woman has ever been screened for
cervical cancer. Previous studies have reported associations
between male-partner communication and women’s health
decisions regarding contraceptive use, family planning and
HIV testing [24–26]. While several small qualitative
studies have recently suggested that male-partner opinions
may be a barrier to cancer screening, especially among
Hispanics and recent immigrants [17–19], this association
has not been previously validated in quantitative fashion.
These studies cited poor knowledge of cervical cancer risk
factors, HPV transmission and treatment, and cultural
ideals of Machismo as potential barriers to cervical cancer
screening. Even after adjusting for all possible confounders
in multivariate analysis, women in our study who believed
that their male partners want them to receive regular Pap
screening were significantly less likely to report ‘never-
screening’. This indicates that perceptions of support from
a woman’s male partner are at least as important for
determining her decision to receive the Pap test indepen-
dent as other economic, psychosocial and health system
barriers to screening.
Given previous reports of the importance of health
access in improving cervical screening participation among
underserved women [27,28], we expected the absence of
insurance/cost barriers to preventive care in our study
setting to translate into above-average rates of screening
participation. However, we found that nearly 11 % of study
participants had never received the Pap test, despite *4
clinic visits in the preceding year. This translates into, a
prevalence of never-screening for our study population that
lies well within the 10–39 % range reported by other U.S.
studies [17,29]. Thus, enrollment of medically under-
served women in a community owned health system was
clearly inadequate to eliminate a key health disparity
despite frequently contact with health care providers. The
fact that the number of health care visits was not identified
as an independent predictor of cervical cancer screening
further confirms that the nature of patient interactions with
their physician as well as other non-economic influences
such as social support, personal beliefs and situational
barriers exert equal or even stronger influences on the
screening behavior than suggested by some earlier reports.
This suggests that full implementation of the ACA will not
automatically translate into increased rates of cervical
cancer screening without concerted efforts to address
health disparities independent of health access. We believe
that our results make a strong case for a ‘‘whole woman’’
approach to improve cervical cancer care, as proposed by
the National Cancer Institute’s Center to Reduce Cancer
Table 3 Reproductive history, social support and Pap test use
Pap never
Yes
N=106
No
N=877
p
Reproductive/medical history
Previous pregnancies, mean (SD) 2.65 ±2.45 3.04 ±1.80 0.14
Previous births, mean (SD 2.07 ±2.55 2.47 ±1.57 0.24
Previous elective pregnancy terminations, n (% yes) 19 (35.9) 187 (41) 0.47
Previous STI, n (% yes) 5 (5) 108 (12.7) 0.02
Contraceptive use, n (% yes) 12 (12) 169 (20.5) 0.04
Does male partner want respondent to have Pap smear? n (% yes) 15 (21.1) 426 (59.4) <0.0001
Do friends and family members receive Pap smear regularly? n (% yes) 20 (22) 438 (53.9) 0.0001
Statistically significant pvalues are highlighted in bold
Unless otherwise stated, data are number and (%) of women answering ‘‘yes’’
1408 J Immigrant Minority Health (2016) 18:1404–1412
123

Health Disparities (CRCHD) [30]. The CRCHD has
emphasized the need to broaden the definition of medical
access beyond monetary support for screening services to
include cultural and educational barriers commonly found
in geographic regions with low rates of screening.
To date, national policy mandates, such as the ACA or
the National Breast and Cervical Cancer Early Detection
Program (NBCCEDP) have focused mainly on reducing
patient entry expense. Neither of these programs supports a
concerted national effort to culturally incentivize patients
to seek care by elimination insidious socio-cultural barri-
ers. One option that could be used to address the barriers of
male partner communication identified here is the use of
female lay health advisors (LHA) to lead community
educational sessions. Implementation of LHA-led pro-
grams have been previously shown to increase rates of
cervical cancer screening by 18–28 % [31,32].
Lastly, our data highlights an ongoing need to appreciate
the socio-cultural and geographical diversity of under-
served populations, and the impact of non-economic dif-
ferences between underserved populations on health
decisions. Medically underserved sub-populations often
have very little in common with one another in terms of
culture, beliefs, group behavior and dynamics. For
Table 4 Impact of knowledge and attitudes on Pap test use
Pap never
Yes
N=106
No
N=877
p
Knowledge, n (% yes)
Pap test is a test that screens for cervical cancer (% true) 54 (87.1) 662 (89.6) 0.54
Abnormal Pap smear can be treated by a doctor (% true) 43 (74.1) 630 (84.1) 0.049
Cervical cancer can be prevented by doing regular Pap smear (% true) 41 (68.3) 598 (79.1) 0.05
I need a Pap smear only when I experience vaginal bleeding other than menses (% true) 7 (11.5) 57 (7.6) 0.29
Factors which increase cervical cancer risk, n (% yes)
Having multiple sexual partners 48 (46.2) 484 (55.4) 0.07
Having sex with a man who has multiple partners 45 (43.3) 448 (51.1) 0.13
Having a history of sexually transmitted disease 40 (38.5) 449 (51.3) 0.01
Giving birth to many children 34 (32.7) 507 (57.8) <0.0001
Family history 41 (39.4) 468 (53.4) 0.007
Poor hygiene 20 (19.2) 252 (28.8) 0.04
Situational barriers [range 1 (definitely yes)–4 (definitely no)], mean (SD)
Pap smear test is too expensive 2.53 ±1.04 2.83 ±1.13 0.02
I find it difficult to communicate with my provider 2.97 ±1.11 3.25 ±1.03 0.02
Screening visits are too long 2.49 ±1.14 3.06 ±1.07 <0.0001
I have no one to care for my other children 3.03 ±1.23 3.47 ±0.94 0.002
It is difficult to get an appointment from the clinic 2.67 ±1.30 2.87 ±1.22 0.19
Attitudes [range 1 (strongly agree)–4 (strongly disagree)], mean (SD)
Perceived personal cervical cancer risk 2.74 ±1.5 2.47 ±1.4 0.67
Cancer risk depends on God’s will 2.51 ±1.24 2.95 ±1.23 0.003
Belief that cancer is incurable 3.24 ±0.96 3.16 ±1.02 0.57
Don’t want to be informed of cancer diagnosis 3.03 ±1.27 3.27 ±1.16 0.10
Want health provider to inform family of cancer diagnosis 1.87 ±1.24 1.96 ±1.19 0.31
Afraid to find out cancer diagnosis 2.24 ±1.30 2.27 ±1.26 0.78
Would be afraid to tell partner of cancer diagnosis or received Pap test 2.84 ±1.34 3.21 ±1.19 0.01
Don’t need Pap smear if I am not sexually active 2.90 ±1.24 3.62 ±0.85 <0.0001
The procedure is painful 2.69 ±1.09 2.95 ±1.07 0.05
Statistically significant pvalues are highlighted in bold
Data are mean ±SD or number (%) of women answering ‘‘yes’’
Score for knowledge was derived by adding up responses to questions which assessed participants’ knowledge of cervical cancer risk and Pap test
benefits
pvalue is based on Wilcoxon rank sum test
J Immigrant Minority Health (2016) 18:1404–1412 1409
123

example, previously reported associations between His-
panic ethnicity, birth outside the U.S., lack of care conti-
nuity, fatalism and beliefs that the ‘‘Pap test is not needed
in women that are not sexually active’’, ‘‘Pap test is pain-
ful’’ and failure to participate in cervical cancer screening
were also observed in our study population [7,13,33,34].
However, other factors potentially influencing screening
behaviors, such as such as fewer years of education and
marital status [29,35,36] were not associated with never
screening in our analyses. These differences may simply
reflect increased awareness as a result of public education
on cervical cancer, or could even be unique to the sub-
population of medically underserved women included in
our study.
One curious finding reported here is that the perception
that ‘‘screening tests are too expensive’’ is more common
among never screeners, despite the absence of an
association between receipt of Medicare and never
screening, and the fact that costs of cervical cancer
screening are heavily subsidized in HHS clinics. The
absence of an association with insurance status is not sur-
prising, considering the safety-net function of HHS.
However, it is possible that responses to these questions
reflect experiences of our subjects outside of the study
venue or that many never screeners base their avoidance of
testing on previous experiences or hearsay. It is not pos-
sible for us to distinguish between possible explanations of
these observations. Nonetheless, we believe that this is a
topic that should be investigated further.
A major strength of this study is the large sample size
and ethnic distribution of the surveyed population. The
robust representation of both medically underserved and
Hispanic women among the study population provides
insight into possible non-access factors among U.S.
Table 5 Odds ratios and 95 % CI of factors associated with Pap ‘never screening’
Factor Adjusted OR, 95 % CI
a
p
Physician/health system influences
Sees same physician (vs. different physician) 0.68 (0.44–1.04) 0.07
Being informed of need for Pap test by a physician (vs. not informed) 0.17 (0.11–0.31) <0.001
Number of health visits 0.95 (0.87–1.02) 0.17
Medical history
Hypertension (vs. no hypertension) 0.68 (0.37–1.22) 0.21
Depression (vs. no) 0.65 (0.30–1.24) 0.22
Previous STI (vs. no STI) 0.38 (0.13–0.87) 0.04
Social influence
Perception of partner support for screening (vs. no/he doesn’t care) 0.06 (0.03–0.14) <0.001
Friends and family members have regular screening (vs. no) 0.37 (0.15–0.99) 0.04
Knowledge
Abnormal test can be treated by doctor (vs. false) 0.53 (0.29–1.03) 0.05
I need Pap only when I experience vaginal bleeding, other than menses (vs. false) 0.63 (0.35–1.18) 0.13
A history of sexually transmitted infection increases risk of cervical cancer (vs. false) 0.57 (0.37–0.87) 0.01
Not getting Pap test increases cervical cancer risk (vs. false) 0.37 (0.24–0.58) <0.0001
Poor hygiene increases cervical cancer risk (vs. false) 0.55 (0.32–0.90) 0.02
Family history of cervical cancer increases risk of disease (vs. false) 0.58 (0.38–0.89) 0.01
Attitude
Cancer risk depends on God’s will (vs. disagree) 1.87 (1.13–3.09) 0.02
Would be afraid to tell partner of cancer (vs. disagree) 1.48 (0.86–2.49) 0.14
Don’t need Pap smear if I am not sexually active (vs. disagree) 3.64 (2.02–6.41) <0.0001
Pap test is painful (vs. disagree) 1.40 (0.84–2.34) 0.20
Situational barriers
Pap smear test is too expensive (vs. not expensive) 1.75 (1.07–2.88) 0.03
I find it difficult to communicate with my provider (vs. not difficult) 1.85 (1.09–3.09) 0.02
Screening visits are too long (vs. not too long) 2.51 (1.53–4.12) 0.0003
I have no one to care for my other children (vs. I have someone) 2.26 (1.27–3.93) 0.004
Statistically significant pvalues are highlighted in bold
a
Each association is adjusted for age, ethnicity, and birth country
1410 J Immigrant Minority Health (2016) 18:1404–1412
123

populations that are at greatest risk for developing cervical
cancer. In addition, population demographics of Harris
County currently match what the racial and ethnic demo-
graphics predicted for the U.S. in 20–30 years [33]. Thus,
our results are only likely to become more relevant to
efforts to eradicate cervical cancer as time passes. Limi-
tations include the fact that all surveys were self-com-
pleted, creating the possibility of self-reporting bias.
Women have been known to report pelvic examinations as
Pap test receipt, thereby inflating screening rates [34,37].
The possibility of selection bias also cannot be ruled out, as
many of the survey non-responders may have been illiter-
ate. To minimize the effect of this limitation, we provided a
bilingual study coordinator to answer participants’
questions.
New Contribution to the Literature
This study demonstrates the impact of male partners on a
woman’s decision to participate in cervical cancer
screening. It also explores reasons why women with fre-
quent contact with the health system may choose never to
participate in screening.
Acknowledgments The authors acknowledge the support of the
Cancer Prevention and Research Institute of Texas (CPRIT PP120091
and PP100201).
References
1. Centers for Disease Control and Prevention (CDC). Cervical
cancer screening among women aged 18–30 years-United States,
2000–2010. MMWR Morb Mortal Wkly Rep. 2013;61:1038–42.
2. Janerich DT, Hadjimichael O, Schwartz PE, et al. The screening
histories of women with invasive cervical cancer, Connecticut.
Am J Public Health. 1995;85:791–4.
3. Singleton H, Orr J. Screening in cancer of the cervix. Philadel-
phia, PA: Lippincott-Raven; 1995.
4. Bitler MP, Carpenter CS. Effects of state cervical cancer insur-
ance mandates on Pap test rates. SSRN 1986270. 2012.
5. Harlan LC, Bernstein AB, Kessler LG. Cervical cancer screening:
who is not screened and why? Am J Public Health. 1991;81:885–90.
6. Chen H, Kessler CL, MoriN, Chauhan SP. Cervicalcancer screening
in the united states, 1993–2010: characteristics of women who are
never screened. J Womens Health. 2012;21:1132–8.
7. Kim K, Elena S, Chen EH, Kim J, Kaufman M, Purkiss J. Cer-
vical cancer screening knowledge and practices among Korean–
American women. Cancer Nurs. 1999;22:297–302.
8. Taylor VM, Yasui Y, Burke N, et al. Pap testing adherence
among Vietnamese American women. Cancer Epidemiol Bio-
mark Prev. 2004;13:613–9.
Table 6 Multivariate analysis of factors associated with never screening
Associated factors Adjusted OR, 95 % CI
a
p
Physician influence
Being informed of need for Pap test by a physician (vs. not informed) 0.34 (0.14–0.80) 0.02
Medical history
Previous STI (vs. no STI) 0.42 (0.06–1.60) 0.27
Partner influence
Partner support (vs. no/he doesn’t care) 0.29 (0.11–0.68) 0.006
Knowledge
Abnormal test can be treated by doctor (vs. false) 0.56 (0.22–1.59) 0.25
A history of sexually transmitted infection increases risk of cervical cancer (vs. false) 1.31 (0.54–3.39) 0.55
Not getting Pap test increases cervical cancer risk (vs. false) 0.87 (0.36–2.20) 0.76
Poor hygiene increases cervical cancer risk (vs. false) 0.70 (0.27–1.71) 0.45
Family history of cervical cancer increases risk of disease (vs. false) 1.10 (0.45–2.76) 0.84
Attitude
Cancer risk depends on God’s will (vs. disagree) 1.27 (0.53–3.00) 0.59
Don’t need Pap smear if I am not sexually active (vs. disagree) 2.93 (0.92–8.57) 0.06
Situational barriers
Pap smear test is too expensive (vs. no) 1.17 (0.49–2.77) 0.72
Screening visits are too long (vs. no) 2.53 (1.06–6.07) 0.04
b
I have no one to care for my other children (vs. no) 1.41(0.49–3.68) 0.5
Statistically significant pvalues are highlighted in bold
a
Each association is adjusted for age, ethnicity, birth country and all other variable in the table
b
Due to the high correlation between ‘‘screening visits are too long’’ and ‘‘difficulty in communication’’, only ‘‘screening visits are too long’’
was used in model
J Immigrant Minority Health (2016) 18:1404–1412 1411
123

9. Wu ZH, Black SA, Markides KS. Prevalence and associated
factors of cancer screening: why are so many older Mexican
American women never screened? Prev Med. 2001;33:268–73.
10. Weinrich S, Coker AL, Weinrich M, Eleazer GP, Greene FL.
Predictors of Pap smear screening in socioeconomically disad-
vantaged elderly women. J Am Geriatr Soc. 1995;43:267–70.
11. Hewitt M, Devesa SS, Breen N. Cervical cancer screening among
US women: analyses of the 2000 national health interview sur-
vey. Prev Med. 2004;39:270–8.
12. Spurlock C, Nadel M, McManmon E. Age and Pap smear history
as a basis for intervention strategy. J Community Health.
1992;17:97–107.
13. Lim AW, Sasieni P. Consultation rates in cervical screening non-
attenders: opportunities to increase screening uptake in GP pri-
mary care. J Med Screen. 2015;22(2):93–9.
14. Cassel J. An epidemiological perspective of psychosocial factors
in disease etiology. Am J Public Health. 1974;64:1040–3.
15. Caplan G. Support systems and community mental health: lec-
tures on concept development. New York: Behavioral Publica-
tions; 1974.
16. Kang SH, Bloom JR. Social support and cancer screening among
older black Americans. J Natl Cancer Inst. 1993;85:737–42.
17. Byrd TL, Chavez R, Wilson KM. Barriers and facilitators of
cervical cancer screening among Hispanic women. Ethn Dis.
2007;17:129.
18. Fernandez ME, McCurdy SA, Arvey SR, et al. HPV knowledge,
attitudes, and cultural beliefs among Hispanic men and women
living on the Texas-Mexico border. Ethn Health. 2009;14:
607–24.
19. Gregg J, Centurion T, Aguillon R, Maldonado J, Celaya-Alston
R. Beliefs about the pap smear among mexican immigrants.
J Immigr Minor Health. 2011;13:899–905.
20. United States Census Bureau. Health insurance coverage at the
county and state level. 2014. http://www.governing.com/gov-
data/health/2009-small-area-health-insurance-coverage-estimates.
html.
21. Harris health system. 2014. https://www.harrishealth.org/en/
news/pages/introducing-harris-health-system.aspx.
22. Eligibility requirements-harris health system. 2014. https://www.
harrishealth.org/en/patients/access-care/eligibility-card/pages/
eligibility-requirements.aspx.
23. National Center for Chronic Disease Prevention and Health
Promotion. CDC behavioral risk factor surveillance system:
prevalence and trends data, nationwide (states, DC, and territo-
ries)-2012 women’s health. 2014.
24. Harvey SM, Henderson JT, Casillas A. Factors associated with
effective contraceptive use among a sample of Latina women.
Women Health. 2006;43:1–16.
25. Sable MR, Campbell JD, Schwarz LR, Brandt J, Dannerbeck A.
Male Hispanic immigrants talk about family planning. J Health
Care Poor Underserved. 2006;17:386–99.
26. Hernandez AM, Zule WA, Karg RS, Browne FA, Wechsberg
WM. Factors that influence HIV risk among Hispanic female
immigrants and their implications for HIV prevention interven-
tions. Int J Family Med. 2012. doi:10.1155/2012/876381.
27. Freeman HP, Wingrove BK. Excess cervical cancer mortality: a
marker for low access to health care in poor communities. 2005,
NIH Pub. No. 05-5282.
28. De Alba I, Hubbell FA, McMullin JM, Sweningson JM, Saitz R.
Impact of U.S. citizenship status on cancer screening among
immigrant women. J Gen Intern Med. 2005;20:290–6.
29. Behbakht K, Lynch A, Teal S, Degeest K, Massad S. Social and
cultural barriers to Papanicolaou test screening in an urban
population. Obstet Gynecol. 2004;104:1355–61.
30. Freeman HP, Wingrove BK. Excess cervical cancer mortality: a
marker for low access to health care in poor communities.
Rockville, MD: National Cancer Institute, Center to Reduce
Cancer Health Disparities, May 2005. NIH Pub. No. 05-5282.
31. Boucher F, Scheniker MB. Cervical cancer among Hispanic
women: assessing the impact on farmworkers. J Immigr Health.
2002;4(3):159–65.
32. Goldsmith DF, Sisneros GC. Cancer prevention strategies among
California farmworkers: preliminary findings. J Rural Health.
1996;12(4):343–8.
33. Day JC. Population projections of the united states, by age, sex,
race, and Hispanic origin: 1992 to 2050. U.S. Department of
Commerce, Economics and Statistics Administration, Bureau of
the Census; 1992.
34. Sawyer JA, Earp JA, Fletcher RH, Daye FF, Wynn TM. Accuracy
of women’s self-report of their last Pap smear. Am J Public
Health. 1989;79:1036–7.
35. Calle EE, Flanders WD, Thun MJ, Martin LM. Demographic
predictors of mammography and Pap smear screening in U.S.
women. Am J Public Health. 1993;83:53–60.
36. Yi JK. Acculturation and pap smear screening practices among
college-aged Vietnamese women in the United States. Cancer
Nurs. 1998;21:335–41.
37. Fruchter RG, Rones K, Roth T, Webber CA, Camilien L, Boyce
JG. Pap smear histories in a medical clinic: accuracy of patients’
self-reports. N Y State J Med. 1992;92:421–4.
1412 J Immigrant Minority Health (2016) 18:1404–1412
123