7: No. 3, May 2010
Factors Associated With
Cervical Cancer Screening in Puerto Rico
Ana P. Ortiz, PhD, MPH; Sarah Hebl; Ruby Serrano, MS, DrPH; María E. Fernandez, PhD; Erick Suárez, PhD; Guillermo Tortolero-Luna, MD, PhD
Suggested citation for this article: Ortiz AP, Hebl S, Serrano R, Fernandez ME, Suárez E, Tortolero-Luna G. Factors associated with cervical cancer screening in Puerto Rico.
[Erratum appears in Prev Chronic Dis 2010;7(5):A58. http://www.cdc.gov/pcd/issues/2010/sep/10_0138.htm.] Prev Chronic Dis 2010;7(3).
may/09_0123.htm. Accessed [date].
Racial/ethnic disparities in cervical cancer screening exist in the United States; rates
are lowest among women who live in Puerto Rico. We identified factors associated with cervical cancer screening among women aged 18 years
or older living in Puerto Rico.
We included women who participated in the Puerto Rico Behavioral Risk Factor Surveillance System
who had not had a hysterectomy (n = 2,206). We calculated the weighted population prevalence estimates
of Papanicolaou (Pap) test screening in the past 3 years and used logistic regression models to assess factors associated with screening.
Most participants (71.9% [95% confidence interval (CI) = 69.4%-74.4%]) reported having had a Pap test in the preceding 3 years. Factors associated with screening in multivariate analysis included routine checkup in the past year and leisure-time physical activity. Compared with women with a household income
less than $15,000, those with higher incomes were more likely to have had a Pap test. Similarly, divorced or separated women were more likely to have been screened (odds
ratio [OR] = 1.13; 95% CI = 1.12-1.15) than those who were married/living together. We did not find associations between screening behavior and education, health care coverage, body mass index, or smoking status.
The prevalence of cervical cancer screening in Puerto Rico is below the 90% recommendation established by Healthy People 2010. Our findings regarding factors associated with
Pap screening behavior identified population subgroups
who are underscreened and who may benefit from targeted interventions and screening programs.
Back to top
In the United States, the incidence of cervical cancer among Hispanics (14.2
per 100,000) is
almost double that of non-Hispanic whites (7.3 per 100,000); the death rate for Hispanics
(3.4 per 100,000) is also
50% higher than for non-Hispanic whites (2.3 per 100,000) (1). Despite the overall decline in
cervical cancer incidence and deaths in the United States in the last few decades, Hispanic women are
less likely to be diagnosed with localized disease and have poorer survival rates than
non-Hispanic whites (1,2). Cervical cancer is the 6th most common cancer among Hispanic women and ranks only 13th among cancers for non-Hispanic whites (2). In Puerto Rico, cervical cancer is the fourth most commonly diagnosed
cancer in women, accounting for 4% of all newly diagnosed cancers and 2% of all cancer-related deaths among women (3).
The use of the Papanicolaou (Pap) test has resulted in a substantial decline
in cervical cancer illnesses and deaths over recent decades (2). In the United
States, lack of cervical cancer screening is the most powerful predictor of cervical cancer; thus, disparities
in Pap test coverage influence disparities in cervical cancer (4). Despite
the accessibility of this screening method, racial/ethnic disparities exist in
its use in the United States (5). For example, in 2004, women living in Puerto Rico had the lowest
having had a recent (last 3 years) Pap test
(73%) when compared with
non-Hispanic whites (87%), African Americans (89%), and Hispanics (87%) living in the United States
(excluding US territories) (6). Among Hispanics and other racial/ethnic groups in the United States, sociodemographic factors such as older age, lower income,
lower education level (particularly those who did not graduate from high school), and lack of health care coverage have been positively correlated with a lower
rate of Pap screening (7,8). In addition to sociodemographic factors, psychosocial factors
such as embarrassment,
language barriers, fear, lack of knowledge, and perceived partner disapproval influence
Hispanic women’s use of Pap screening (9).
To the best of our knowledge, no study has explored the factors influencing Pap test
use among women in Puerto Rico. The purpose of this study was to determine the factors associated with self-reported recent Pap test (within 3 years before the interview) among women aged 18 years
or older living in Puerto Rico who participated in the Puerto Rico Behavioral Risk Factor Surveillance System (PR-BRFSS). The results of our study are needed to elucidate barriers to cervical cancer screening in
the Hispanic population of Puerto Rico. Our objectives are in alignment with the National Breast and Cervical Cancer Early Detection Program’s goal of reducing racial disparities in screening and early detection and with the Healthy People 2010 goal of reducing health disparities (10). In addition, this study contributes to
meeting the objective of the Puerto Rico Comprehensive Cancer Control Plan (11) of increasing the proportion of women aged 18 years or older
who receive a
Pap test consistent with current recommendations.
Back to top
For this study, we used 2006 data from the PR-BRFSS (12). The PR-BRFSS is part of the national BRFSS, a state-based system of health surveys established in 1984 by the Centers for Disease Control and Prevention to collect information on health risk behaviors, preventive health practices, and health care access primarily related to chronic disease and injury. This cross-sectional survey is conducted annually among noninstitutionalized adults,
aged 18 years or older, in all 50 states, the
District of Columbia, Puerto Rico, the US Virgin Islands, and Guam. The institutional
review board of the University of Puerto Rico Medical Sciences Campus approved this study.
The women eligible for this study had participated in the 2006 PR-BRFSS, were
aged 18 years or older, had no history of hysterectomy, and had responded to the question of whether or not they
had had a Pap
test in the past 3 years before the interview. Among the 3,040 women aged 18
years or older who participated in the PR-BRFSS in 2006, we excluded 710 women who had had a hysterectomy and 124 women with missing information regarding age (n = 16), Pap test history (n = 38), time since
last Pap test (n = 31), or hysterectomy status (n = 39), which left a final study
sample of 2,206 women.
The outcome variable of interest for this analysis was the proportion of women who had had a Pap test in the 3 years before the interview. Factors of interest included demographic characteristics, such as age in years (18-20, 21-29, 30-39, 40-49, 50-64,
≥65), marital status (married/living together, divorced/separated, widowed, single), educational attainment (less than high school graduate, high school graduate/General
Educational Development certification, some college/technical school, college graduate), household income
(<$15,000, $15,000-$34,999, $35,000-$49,999, ≥$50,000), health care coverage (yes/no), employment status (currently employed, unemployed, homemaker or retired, unable to work, student), and number of children living in the household (0, 1, 2, ≥3). Clinical characteristics included body mass index (BMI, categorized as normal or underweight [18.5-24.9 kg/m2], overweight [25.0-29.9 kg/m2],
or obese [≥30.0 kg/m2]), routine
checkup in the past year (yes/no), and perceived general health status (fair or poor, good
Lifestyle characteristics included current smoking status (yes/no), binge drinking (4 or more drinks on 1 occasion [yes/no]), and leisure-time physical activity in the past 30 days (yes/no).
We conducted the statistical analysis by using SAS version 9.2 (SAS Institute, Inc, Cary, North Carolina) and Stata version 10.0 (StataCorp LP, College Station, Texas). We first described the study
sample according to demographic, clinical, and lifestyle characteristics by using the survey frequency function in SAS. We then assessed the relationship between cervical cancer screening behavior and the demographic, clinical, and lifestyle factors by using contingency tables and
Pearson’s χ2 tests, which also required the use of the survey frequency function. To further assess these relationships, we used the
generalized linear model procedure in Stata to construct simple and multivariable logistic regression models (13). We estimated the prevalence odds ratios and their 95% confidence intervals to determine the magnitude of the association between the specific factors and cervical cancer screening behavior. The variables
with cervical cancer screening (P < .05) in the age-adjusted logistic regression models were included in the multivariable logistic regression models; those with at least marginal
significance (P < .10) in the multivariate model were retained in the model. All data were weighted according to the respondent’s age and the inverse of
her probability of selection by using the 2006
census population projections. Detailed information about BRFSS weighting procedures can
be found in the BRFSS operational guide (14).
Back to top
Approximately half of the women
were aged 39 years or younger and were married or living as a couple
(Table 1). Most had a household income of less than $35,000 a year, and
almost all reported having health care coverage. Two-thirds of women reported that
they were in good to excellent general health; 81% had a routine checkup in the previous year, 82% had at least 1
Pap test in their lifetime, and 72% had a Pap
test in the past 3 years.
In the bivariate analysis, Pap screening in the past 3 years was significantly associated
(P < .05) with age, marital status, education level, household income, employment status, number of children in the household,
BMI, routine physical examination in the past year, health status, and leisure-time physical activity. Lower
rates of Pap screening were observed in younger women, single
women, women with a household income less than $15,000, women with some college or technical school
education, and students (Table 2). In addition, lower rates of Pap test screening were reported by
underweight or normal-weight women, women who had not had a routine medical checkup in the prior year, and those who reported no leisure-time physical activity.
Pap screening was not associated with smoking status, health care coverage, binge drinking, or heavy
We found no significant interactions in the multivariate model
(likelihood ratio χ2 =
70.13, P = .17). Compared with younger women (aged 18-20 years), older women (aged ≥21 years) were 2
to 5 times as likely to have had a Pap test in the last 3 years
(Table 3). In addition,
compared with women with the lowest household income, women with larger
household incomes were also more likely to have had a Pap test in the past 3
years. Single women (never married) and widows were less likely than married
women to have had a Pap test in the past 3 years.
Back to top
In 2006, Puerto Rico fell short of meeting the Healthy People 2010 goal of a 90% Pap test screening rate for women aged 18 years or older (15). Our study shows that younger (aged
<30 years) and older (aged
≥65 years) Puerto Rican women reported lower rates of cervical cancer screening than women aged 30
to 64 years. These results are consistent with the overall screening patterns in the United States (16) and among Hispanic women in the
rest of the United States (8). Of interest is the
low prevalence of screening among women aged 18
to 20 years in Puerto Rico observed in our study (9%). This prevalence is consistent with the low rate of cervical cancer screening among women aged 18
to 24 years in Puerto Rico (41%) reported by the BRFSS for 2004, which in
addition is much lower than the screening rate of their counterparts in the rest
of the United States (median = 81%), a pattern that has been consistent since 1996 (6). In addition, women with a household income of
at least $35,000 were 2 to 3 times as likely to have been screened as those with the lowest household income. Household income has been positively correlated with cervical cancer screening in multiple studies (10,16-17), an association that could be partially explained by improved access to care with increasing wealth.
In contrast to results in other populations, our study found no substantial positive effect of education, employment status, or health care coverage on screening practices (7,8,16,17). Although the reasons for the lack of association between screening behavior and education and employment are unknown, for health care coverage the lack of association may be explained by the high prevalence of women in our study who had health care coverage (94%). The
health care reform legislation passed
in Puerto Rico in 1993 made health insurance available for underserved populations in Puerto Rico at or below 200%
of the poverty level.
Routine checkup in the past year increased the likelihood of having been
screened, a result consistent with other studies (8,18). This finding is not
surprising since cervical cancer screening is often recommended by the doctor
during the clinical visit. Participating in leisure-time physical activity
increased the likelihood of screening in our study by more than 60%, a result
consistent with that of another recent study that found that exercise was positively
associated with breast and cervical cancer screening (7). This previous study
also found a positive association between nonsmoking and screening. Our
research, however, did not observe this association. Although studies have consistently shown an
inverse relationship between obesity and screening behavior (decreased cervical
cancer screening with increasing body size) (19), we found no association
between BMI and cervical cancer screening in multivariate analysis. While in
some groups higher weight may be associated with less emphasis on health and
thus less screening, our findings may reflect a cultural norm of acceptance of a
larger body weight/body size in Puerto Rican women similar to that found is some
populations of black women in the United States (20).
Our study is subject to limitations. First, because the BRFSS is a telephone-based survey, it includes data
only from residents
who have a working home telephone and, thus, is unable to survey those who reside in households without telephone access. Consequently, the above data may not be generalizable to the entire adult Puerto Rican female population. Evidence suggests that income is positively associated with
a recent Pap test. Since women who do not have telephones are more likely
to have lower incomes, they likely also have lower Pap screening rates than
women who participated in the study (8). Nonetheless, response rates in the 2006 PR-BRFSS were much higher (74%) than the median reported for BRFSS surveys in all other US states and territories
of the United States for the same year (51%) (21). Finally, the prevalence estimates were based on self-reported information, which is subject to recall bias and social desirability bias.
Our study shows that the rates of cervical cancer screening in Puerto Rico
are lower than those in the United States, and well below the 90% goal established by Healthy People
2010. We have identified factors associated with Pap test screening
in Puerto Rico that help identify subgroups of the population who are underscreened
and who would benefit from targeted interventions. Interventions should focus on increasing screening rates, particularly among young and low-income women. Future studies should also focus
on other psychosocial correlates of screening, including attitudes, beliefs, and cultural norms regarding screening practices. Studies of the
effect of social determinants of health, such as area of residence (rural vs urban) and migrant status, are warranted in this population,
since these factors influence screening behavior (17,22). In addition, public health policy regarding universal mandatory cervical cancer screening coverage in Puerto Rico should be implemented.
Back to top
This work was funded by the following grants: National Cancer Institute (U54CA96297 and UO1CA86117-01), National
Center for Research Resources (G12RR03051), and Centers for Disease Control
and Prevention (5U58DP222931-05), and by an unrestricted educational grant
from Bristol-Myers Squibb, Puerto Rico. We also acknowledge the support from the Biostatistics and Bioinformatics Core of the U54 grant, a partnership
between the University of Puerto Rico and MD Anderson Cancer Center (NIH
NCI U54CA96297) for the review of this manuscript.
Back to top
Corresponding Author: Ana P. Ortiz, PhD, MPH, University of Puerto Rico Comprehensive Cancer Center
and University of Puerto Rico Medical Sciences Campus, PMB 711, 89 De Diego Ave, Ste 105, San Juan, Puerto Rico 00927-6346. Telephone: 787-758-2525, ext 1471. E-mail:
Author Affiliations: Sarah Hebl, Tufts University School of Medicine, Boston,
Massachusetts; Ruby Serrano, Puerto Rico Behavioral Risk Factor Surveillance
System, Puerto Rico Department of Health, San Juan, Puerto Rico; María E. Fernandez, University of Texas at Houston, Houston, Texas; Erick Suárez, University of Puerto Rico Medical Sciences Campus, San Juan, Puerto Rico; Guillermo Tortolero-Luna, University of Puerto Rico Comprehensive Cancer Center, San Juan, Puerto Rico.
Back to top
- Cancer facts and figures for Hispanics, 2006-2008. Atlanta (GA): American Cancer Society; 2006. http://www.cancer.org/downloads/STT/CAFF2006HispPWSecured.pdf. Accessed June 18, 2009.
- Howe HL, Wu X, Ries LA, Cokkinides V, Ahmed F, Jemal A, et al.
Annual report to the nation on the status of cancer, 1975-2003, featuring cancer among U.S. Hispanic/Latino populations. Cancer 2006;107(8):1711-42.
- Cancer of the cervix uteri stat fact sheet, Puerto Rico Central Cancer Registry. Puerto Rico Department of Health; 2003. http://www.salud.gov.pr/RCancer/Reports/Pages/default.aspx. Accessed June 18, 2009.
- Akers AY, Newmann SJ, Smith JS.
Factors underlying disparities in cervical cancer incidence, screening, and treatment in the United States. Curr Probl Cancer 2007;31(3):157-81.
- Smith RA, Cokkinides V, Brawley OW.
Cancer screening in the United States, 2009: a review of current American Cancer Society guidelines and issues in cancer screening. CA Cancer J Clin 2009;59(1):27-41.
- Behavioral Risk Factor Surveillance System survey data. Atlanta (GA): US Department of Health and Human Services, Centers for Disease Control and Prevention.
Accessed May 15, 2007.
- Welch C, Miller CW, James NT.
Sociodemographic and health-related determinants of breast and cervical cancer screening behavior, 2005. J Obstet Gynecol Neonatal Nurs 2008;37(1):51-7.
- Coughlin SS, Uhler RJ.
Breast and cervical cancer screening practices among Hispanic women in the United States and Puerto Rico, 1998-1999. Prev Med 2002;34(2):242-51.
- Fernández ME, Gonzales A, Tortolero-Luna G, Williams J, Saavedra-Embesi M, Chan W, et al.
Effectiveness of cultivando la salud: a breast and cervical cancer screening promotion program for low-income Hispanic women. Am J Public Health 2009;99(5):936-43.
- Adams EK, Breen N, Joski PJ.
Impact of the National Breast and Cervical Cancer Early Detection Program on mammography and
Pap test utilization among white, Hispanic, and African American women: 1996-2000. Cancer 2007;109(2 Suppl):348-58.
- Puerto Rico Comprehensive Cancer Control Plan, 2008-2012. San Juan (PR): Comprehensive Cancer Control Coalition; 2008.
- Behavioral Risk Factor Surveillance System survey data. Atlanta (GA): US
Department of Health and Human Services, Centers for Disease Control and
Prevention. http://apps.nccd.cdc.gov/brfss/page.asp?cat=XX&yr=2006&state=PR#XX. Accessed February 18, 2010.
- Hosmer D, Lemeshow S. Applied logistic regression. 2nd edition. New York (NY): John Wiley & Sons, Inc; 2000. p. 31-46.
- Behavioral Risk Factor Surveillance System operational and user’s guide,
version 3.0. Atlanta (GA): US Department of Health and Human Services,
Centers for Disease Control and Prevention. ftp://ftp.cdc.gov/pub/Data/Brfss/userguide.pdf. Accessed
February 18, 2010.
- Healthy people 2010. Volumes I and II. US Department of Health and Human Services; 2001. http://www.healthypeople.gov/Document/tableofcontents.htm#under. Accessed June 18, 2009.
- Coughlin SS, Leadbetter S, Richards T, Sabatino SA.
Contextual analysis of breast and cervical cancer screening and factors associated with health care access among United States women, 2002. Soc Sci Med 2008;66(2):260-75.
- Coughlin SS, King J, Richards TB, Ekwueme DU.
Cervical cancer screening among women in metropolitan areas of the United States by individual-level and area-based measures of socioeconomic status, 2000 to 2002. Cancer Epidemiol Biomarkers Prev 2006;15(11):2154-9.
- Shah M, Zhu K, Wu H, Potter J.
Hispanic acculturation and utilization of cervical cancer screening in the US. Prev Med 2006;42(2):146-9.
- Cohen SS, Palmieri RT, Nyante SJ, Koralek DO, Kim S,
Bradshaw P, et al.
Obesity and screening for breast, cervical and colorectal cancer in women: a
review. Cancer 2008;112(9):1892-904.
- Anderson LA, Eyler AA, Galuska DA, Brown DR, Brownson RC.
Relationship of satisfaction with body size and trying to lose weight in a
national survey of overweight and obese women aged 40 and older, United
States. Prev Med 2002;35(4):390-6.
- Kilmer G, Roberts H, Hughes E, Li Y, Valluru B, Fan A, et al.
of certain behaviors and conditions among states and selected local areas
— Behavioral Risk Factor Surveillance System (BRFSS), United States, 2006. MMWR Surveill Summ 2008;57(SS-7):1-10.
- Coughlin SS, Wilson KM.
Breast and cervical cancer screening among migrant
and seasonal farmworkers: a review. Cancer Detect Prev 2002;26(3):203-9.
Back to top