Breast and Cervical Cancer Screening Utilization Among Hispanic Women Living Near the United States-Mexico Border

Abstract

Objective:

Hispanic women who reside in low-resource settings are especially at risk for nonparticipation in cancer screening programs. The purpose of this study was to assess characteristics that influence breast and cervical cancer screening among older Hispanic women living along the United States-Mexico border.

Methods:

A cross-sectional study of women aged ≥50 years (n = 504) residing in Yuma County, Arizona, were randomly selected for interviews. Logistic regression analyses were conducted to identify determinants of compliance with mammography and Pap smear use.

Results:

Women who received a recommendation from a clinician to get both mammography and Pap smears were more likely to receive a mammogram within the past year (adjusted odds ratio [AOR] 5.1, 95% confidence interval [CI] 3.0–8.9) compared to women who received no recommendation. Likewise, women who received both recommendations were more likely to receive a Pap smear within the past 3 years (AOR 9.7, 95% CI 4.6–20.7) compared to women who received no recommendation. Other factors, such as current health insurance and a visit with their healthcare provider in the past year, were also associated with getting a mammogram within 1 year or Pap smear within 3 years.

Conclusions:

Enabling characteristics were significantly associated with breast and cervical cancer screening use compared to predisposing and need characteristics among older Hispanic women residing near the U.S.-Mexico border. Clinician recommendation of both mammograms and Pap smears and opportunistic clinic visits to medical providers may increase breast and cervical cancer screening coverage and reduce the burden of these two cancers in this high-risk population.

Introduction

Cancer is the second leading cause of death among Hispanics in the United States.¹ Breast cancer is the most common neoplasm among Hispanic women in terms of incidence (90.2/100,000) and mortality (15.6/100,000).² Despite a lower incidence and mortality compared to non-Hispanic white (NHW) women (126.9/100,000 and 24.4/100,000, respectively),² breast tumors in Hispanics are more likely to appear at a larger size and more advanced stage disease.^2
–4 By comparison, cervical cancer has a higher incidence (12.7 vs. 7.3/100,000) and mortality (3.1 vs. 2.1/100,000) among Hispanic women compared to their NHW counterparts.² Other national data continue to demonstrate a higher cervical cancer incidence among Hispanics (13.9 vs. 7.7/100,000).⁵ This trend is particularly pronounced for foreign-born women living in the United States, who have experienced a 22% increase in cervical cancer mortality rates since 1985.⁶ Along the United States-Mexico border, Hispanic women also have almost twice the cervical cancer incidence (13.9 vs. 7.0/100,000) compared to NHW women.⁷

One major factor contributing to these disparities in cancer rates and outcomes is the underuse of cancer screening services. The best population-based estimate for this comes from the National Health Interview Survey (NHIS) performed by the Centers for Disease Control and Prevention (CDC). Per these estimates, Mexican-origin Hispanic women are less likely to have mammography within the past year (49%, compared to NHW women at 56%) or Pap smear within the past 3 years (73%, compared to NHW women at 80%),² as recommended by the American Cancer Society (ACS). Along the U.S.-Mexico border, these differences may be even more pronounced, with only 55.5% of Hispanic women in Texas who were interviewed in Spanish reporting a Pap smear within the past 3 years,⁸ and 36% of Hispanic women reporting a mammogram within the past 1 year.⁹ A greater understanding of access to and use of cancer screening services is needed specifically for border populations.¹⁰ In recognition of the magnitude of the problem, the U.S.-Mexico Border Health Commission has identified the reduction of breast and cervical cancer mortality as a major objective of the Healthy Border 2010; thus, improved screening rates for these cancers is essential.¹¹

In this study, we examine the use of mammography and cervical cytology (Pap smear) using the framework of the revised Behavioral Model for Vulnerable Populations adapted for use with border populations.^9,12 This model conceptualizes use of healthcare as an outcome determined by the interplay of predisposing (socioeconomic) characteristics, enabling factors, and need factors.¹³ Because vulnerable populations, such as low-income Hispanic women who reside in a low-resource setting such as the U.S.-Mexico border, face additional barriers to healthcare, we also consider health behavior factors for this setting. Our hypothesis is that Hispanic women who are more acculturated, have higher income, and have insurance coverage will be more likely to participate in breast and cervical cancer screening.

The purpose of the present analysis was to describe breast and cervical cancer screening behaviors among postreproductive-age Hispanic (U.S.-born and Mexican-born) women residing in a rural county along the U.S.-Mexico border. The underlying goal was to obtain specific cancer screening use information from a population of women who are at risk of nonparticipation within the current surveillance systems because of cultural and linguistic barriers, low socioeconomic status (SES), and fear of disclosure of nonlegal immigration status.

Materials and Methods

This study provided the baseline information for a randomized controlled trial (RCT) of a behavioral intervention to promote long-term breast and cervical cancer screening, which is to be published separately. Participants were selected from the 7 of 24 county census tracts in Yuma County with >50% Hispanic population. The target area included three small urban centers (Yuma, Somerton, and San Luis) as well as surrounding rural agricultural communities. The entire study data collection and follow-up assessment were completed by December 2006. The study was approved by the Human Subjects Protection Program at the University of Arizona.

Recruitment and population

Dwellings were identified using detailed census tract maps. From the 24 census tracts that comprise Yuma County, U.S. Census Bureau data were used to identify the 7 tracts that had >50% Hispanic population. All dwellings in each of the 7 selected census tracts were identified. Dwellings were assigned sequential numbers and randomly selected using a random number generator to achieve proportional representation of respondents from each of the census tracts. For each identified household, a maximum of three home visits on different days and at varying times was attempted in order to contact potential subjects from each selected household before excluding the household from the survey. Subjects were eligible for inclusion into the study if they were (1) women ≥50 years of age, (2) residents of the household selected, and (3) able to understand either English or Spanish. Exclusion criteria included being (1) male, (2) female under the age of 50, (3) a nonresident of the household selected, and (4) unable to understand either English or Spanish. If more than one woman per household was eligible, the individual whose birthday was nearest the date of the interviewer's visit was invited to participate.

All interviews and subject contacts were performed by female bilingual community health workers (promotoras) who were native to the communities being surveyed and employed by the Western Arizona Area Health Education Center (WAHEC). Interviewers received a total of 12 hours of research project-specific training by study staff during two sessions that occurred before and during the study period and were supervised by experienced field staff on a weekly basis to assure the fidelity and completeness of the structured scripted interviews. The training workshops covered (1) cancer prevention in general, (2) breast and cervical cancer screening guidelines, (3) the purpose of the study and the role of the community health workers in the study, (4) the questionnaires being used in Spanish and English, (5) study protocols for introduction, ascertainment of eligibility, and study consent documents, and (6) issues of confidentiality. All field staff completed Human Subjects Certification Training from the University of Arizona Human Subjects Protection Program.

A total of 1255 contacts were made in order to identify 639 eligible women and enroll 504 participants (79% participation). Of the 135 eligible women who did not enroll in the study, the major reasons cited included lack of time and family care issues. However, there were no differences in age, census tract, or ethnicity between the 504 participants and the 135 nonparticipants (data not shown).

Interviewers had a written script in English or Spanish to introduce themselves at household dwellings. This included their name, their employment affiliation with WAHEC and the University of Arizona, and the purpose of the study. The informed consent form was provided to each potential participant and verbally reviewed with them. Potential subjects were encouraged to ask questions about the study. They were also offered the opportunity to complete the form at a later time. All participants completed the written informed consent document, as well as an interviewer-administered questionnaire in the self-selected preferred language (English or Spanish), and all surveys were completed in the homes of participants.

Survey instrument

The survey instrument used in this study was developed collaboratively with researchers, promotoras, and community stakeholders for use among Hispanic women in southwest border communities and has been described previously.¹⁰ The instrument was adapted to provide insight into the health behaviors of older postreproductive-age Hispanic women. The questionnaire assessed healthcare and preventive services use, including cancer screening. Other items measured sociodemographic information, access to care, prevention orientation, general health, acculturation, language preference, and knowledge, attitudes, and behaviors about breast and cervical cancer and screening.

Statistical analysis

Descriptive statistics were calculated for demographic, use, and risk factor characteristics. The primary outcomes were self-reported use of breast and cervical cancer screening. Variables were modeled to investigate if they were independently associated with breast and cervical cancer screening use. These included demographic, healthcare coverage, access to care, use, sexual history, and acculturation variables. The variables were selected using the framework proposed by the revised Behavioral Model for Vulnerable Populations¹² and adapted for use with border populations.⁹ The independent variables and outcomes included in the analyses are discussed below:

Predisposing factors included country of origin, age, acculturation level, education, employment, marital status, and length of U.S. residency. Acculturation level was coded using the acculturation framework of Balcazar et al.,¹⁴ which classifies acculturation level as low, bicultural, or high. Acculturation level was only asked of Hispanic subjects.

Enabling factors were monthly income, current medical insurance, regular source of medical care, visited healthcare provider in the last 12 months for any reason, and if the woman remembered if a provider had ever given a recommendation for a mammogram or Pap smear.

Need and health behavior factors included self-rated health status, current tobacco use, hysterectomy, and number of pregnancies. Self-rated health was rated on a 4-point scale (excellent, good, average, or poor).

Self-reported screening outcomes used to measure use included mammogram within 1 year and Pap smear within 3 years. For women aged ≥40, the ACS recommends yearly mammograms.¹⁵ For Pap smears, the ACS recommends that beginning at age 30, women who have had three normal Pap test results in a row may be tested less often, every 2–3 years.¹⁵

Logistic regression analyses at the p=0.05 level of significance were conducted to identify variables potentially individually associated with self-reported use of mammography and Pap smears within 1 year and 3 years, respectively. Multivariate logistic regression analyses were performed to model those variables that independently predicted self-reported screening history or could be possible confounders. Odds ratios (OR), adjusted odds ratios (AOR), and 95% confidence intervals (95% CI) were calculated as a measure of the association between determinants and outcomes. All statistical analyses were performed using Stata version 10.0 (College Station, TX).

Results

Questionnaire data were available for all 504 participants who filled out a survey. Mexican-born (74%) and U.S.-born Hispanic (16%) women represented the great majority of the study population, although NHW women (10%) also participated in the survey. All Mexican-born Hispanic women (n = 371) completed the questionnaire in Spanish, and all U.S.-born NHW women (n = 52) completed the questionnaire in English. Among U.S.-born Hispanic women (n = 81), 79% completed the questionnaire in Spanish and 21% completed it in English.

Statistically significant differences existed between Mexican-born Hispanic, U.S.-born Hispanic, and NHW women with regard to distribution of predisposing, enabling, and need variables (Table 1). Mexican-born Hispanic women were younger, less acculturated, had less education, were more likely to be married, had a lower monthly income, were less likely to have health insurance but more likely to have a regular source of care, were less likely to rate their health as excellent or good, and were less likely to have used tobacco ever compared to U.S.-born Hispanic women and NHW women. Although a regular source of care was reported by 90% of respondents, 62% did not seek medical services when they needed them (data not shown). There were no statistically significant differences between Mexican-born Hispanic, U.S.-born Hispanic, and NHW women for current employment, visiting healthcare provider in the past 12 months, and having had a Pap smear within 3 years. Interestingly, provider recommendation for breast and cervical cancer screening and having had a hysterectomy did not statistically differ between Mexican-born and U.S.-born Hispanic women but did differ between those two populations and NHW women. Self-reported screening behaviors also did not statistically differ between Mexican-born and U.S.-born Hispanic women.

Table 1.

Distribution of Characteristics Among Random Sample of Women in Yuma County, by Ethnicity and Place of Birth ( n = 504)

Characteristic	n	%	Mexican-born Hispanic (n = 371) %	U.S.-born Hispanic (n = 81) %	U.S.-born non-Hispanic white (n = 52) %	p value ^a
Predisposing factors
Age, years
50–65	349	69	76	51	50	<0.001
≥66	155	31	24	49	50
Acculturation
Low	314	80	93	21	N/A	<0.001
Bicultural or high	76	20	7	79	N/A
Education
Less than elementary school	202	46	58	27	0	<0.001
Elementary completed to high school completed	199	45	37	65	67
Some college or more	39	9	5	8	33
Currently employed
No	443	88	89	90	82	0.36
Yes	59	12	11	10	18
Married/cohabitating
No	181	38	33	56	44	0.001
Yes	296	62	67	44	56
Residence in United States, years
0–30	272	54	70	12	6	<0.001
≥31	232	46	30	88	94
Enabling factors
Monthly income
$0–$800	230	49	54	41	18	<0.001
≥$801	242	51	46	59	82
Current medical insurance
No	220	44	53	25	4	<0.001
Yes	284	56	47	75	96
Regular source of medical care
No	49	10	7	16	16	0.02
Yes	453	90	93	84	84
Visited healthcare provider in last 12 months
No	74	15	14	20	15	0.62
Yes	425	85	86	80	85
Provider recommendation ever for
Neither Pap smear nor mammogram	135	27	28	28	17	0.01
Mammogram only	28	6	4	4	17
Pap smear only	46	9	10	6	10
Both	292	58	58	62	56
Need and health behaviors
Self-rated health as:
Excellent or good	171	34	26	45	71	<0.001
Average	277	55	63	40	17
Poor	56	11	10	15	12
Used tobacco ever
No	356	72	77	59	54	<0.001
Yes	142	28	23	41	46
Hysterectomy
No	289	58	61	56	37	0.01
Yes	211	42	39	44	63
Screening behaviors
Had mammogram within 1 year (n = 504)	228	45	45	37	62	0.02
Had Pap smear within 3 years^b (n = 289)	212	73	75	64	72	0.33

p value for Pearson's chi-square test of homogeneity.

Pap smears for women who have not had a hysterectomy.

N/A, not available.

The NHW women who participated in the survey were not included in further analyses for this study. We did not include them in any modeling for screening behavior, given the study aims to assess breast and cervical cancer screening behaviors among postreproductive-age Hispanic women.

Table 2 shows results of a stratified analysis to assess possible effect measure modification between Hispanic ethnicity and place of birth. Comparing the 95% CIs of the unadjusted ORs for all predisposing, enabling, and need and health behavior variables, the results show that no effect measure modification was found. As no statistically significant measures of effect were found for any need and health behavior variables, they are excluded from Table 2 but are available upon request. Of note, where measures of effect appear to be strongest, such as provider recommendation ever and visiting provider in the last 12 months, there was no statistically significant difference between those variables for Mexican-born and U.S.-born Hispanic women. Thus, results of crude and adjusted analyses are displayed with the Hispanic sample combined in all further analyses.

Table 2.

Mammography and Pap Smear Screening: Crude Associations with Predisposing and Enabling Variables, Stratified by Place of Birth and Hispanic Ethnicity

	Born in Mexico (n = 371) Mammogram within 1 year		U.S.-born Hispanic (n = 81) Mammogram within 1 year		Born in Mexico (n = 226)^a Pap smear within 3 years		U.S.-born Hispanic (n = 45)^a Pap smear within 3 years
	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Predisposing factors
Age
50–65 (ref)	1.0		1.0		1.0		1.0
≥66	1.5	1.0–2.5	0.4	0.2–1.1	1.1	0.5–2.2	0.3	0-.1–1.2
Acculturation
Low (ref)	1.0		1.0		1.0		1.0
Bicultural or high	0.9	0.4–2.1	2.4	0.6–10.0	1.3	0.4–4.1	3.7	0.9–15.0
Education
Less than elementary school (ref)	1.0		1.0		1.0		1.0
Elementary completed to high school completed	1.0	0.7–1.6	1.4	0.4–4.2	1.2	0.6–2.3	7.0	1.6–30.9
Some college or more	1.1	0.4–3.2	11.7	1.1–122.4	1.7	0.4.-8.4	-	-
Currently employed
No (ref)	1.0		1.0		1.0		1.0
Yes	0.8	0.4–1.6	1.8	0.4–7.8	1.9	0.8–5.1	-	-
Married/cohabitating
No (ref)	1.0		1.0		1.0		1.0
Yes	0.8	0.5–1.3	1.2	0.5–2.9	1.6	0.8–3.1	0.6	0.2–2.2
Years residence in United States
0–30 (ref)	1.0		1.0		1.0		1.0
≥31	1.4	0.9–2.1	2.6	0.5–13.2	0.9	0.5–1.9	1.6	0.4–7.1
Enabling factors
Monthly income
$0–$800 (ref)	1.0		1.0		1.0		1.0
≥801	0.9	0.6–1.4	2.6	0.9–7.4	1.3	0.7–2.4	1.5	0.4–5.3
Current medical insurance
No (ref)	1.0		1.0		1.0		1.0
Yes	2.0	1.3–3.0	3.0	0.9–9.9	1.5	0.8–2.8	0.9	0.2–4.1
Regular source of medical care
No (ref)	1.0		1.0		1.0		1.0
Yes	1.8	0.8–4.1	9.2	1.1–74.5	3.1	1.3–7.8	3.0	0.4–20.2
Visited healthcare provider in last 12 months
No (ref)	1.0		1.0		1.0		1.0
Yes	3.8	1.8–7.9	12.8	1.6–102.8	4.4	1.9–9.9	4.0	0.9–17.6
Provider recommendation ever
Pap smear: No (ref)	1.0		1.0		1.0		1.0
Pap smear: Yes	3.4	2.1–5.4	7.4	2.0–27.5	4.8	2.5–9.2	19.1	3.9–94.0
Mammogram No (ref)	1.0		1.0		1.0		1.0
Mammogram Yes	4.5	2.8–7.2	5.8	1.8–18.9	5.1	2.7–9.8	27.1	5.3–139.6
Neither (ref)	1.0		1.0		1.0		1.0
Mammogram only	1.9	0.6–5.7	-	-	1.8	0.3–10.3	-	-
Pap smear only	0.9	0.4–2.2	1.7	0.1–20.4	1.8	0.6–5.3	5.5	0.5–65.1
Both	4.6	2.7–7.8	7.2	1.9–27.4	6.2	3.1–12.6	44.0	6.4–302.0

Sample size includes only women who have not had a hysterectomy.

CI, confidence interval; OR, odds ratio; ref, referent.

Unadjusted analyses between predisposing and enabling variables with mammography and Pap smear use are presented in Table 3. Enabling variables were found to be the strongest predictors of mammography and Pap smear use for Hispanic women. For example, women who reported having visited a healthcare provider in the last 12 months were more likely to report a mammogram within 1 year and Pap smear within 3 years (OR 4.7, 95% CI 2.4–9.2 and OR 4.4, 95% CI 2.2–9.0, respectively). A result of particular interest is that those who ever received a doctor recommendation for both mammography and Pap smear were more likely to report a mammogram within 1 year and Pap smear within 3 years compared to those who received no recommendation (OR 4.9, 95% CI 3.0–7.9 and OR 8.2, 95% CI 4.3–15.7). In contrast, among predisposing variables, only one variable (currently employed) showed marginal significance for Pap smear within 3 years. Again, as there were no statistically significant associations between any needs variables and screening outcomes, these variables were excluded from Table 3 but are available upon request.

Table 3.

Mammography and Pap Smear Screening: Crude Associations with Predisposing and Enabling Variables Among a Random Sample of Hispanic Women in Yuma County, Arizona

	Mammogram within 1 year (n = 452)		Pap smear within 3 years (n = 271)^a
	OR	95% CI	OR	95% CI
Predisposing factors
Country of Birth
Mexico (ref)	1.0		1.0
United States	0.7	0.4–1.2	0.6	0.3–1.2
Age
50–65 (ref)	1.0		1.0
≥66	1.1	0.7–1.7	0.7	0.4–1.3
Acculturation
Low (ref)	1.0		1.0
Bicultural or high	0.8	0.5–1.4	1.1	0.6–2.4
Education
Less than elementary school (ref)	1.0		1.0
Elementary completed to high school completed	1.0	0.6–1.5	1.5	0.8–2.8
Some college or more	1.8	0.7–4.3	2.3	0.5–11.0
Currently employed
No (ref)	1.0		1.0
Yes	0.9	0.5–1.7	2.5	0.9–6.7
Married/cohabitating
No (ref)	1.0		1.0
Yes	0.9	0.6–1.4	1.4	0.8–2.4
Years residence in United States
0–30 (ref)	1.0		1.0
≥31	1.2	0.8–1.7	0.9	0.5–1.6
Enabling factors
Monthly income
$0–$800 (ref)	1.0		1.0
≥$801 or more	1.0	0.7–1.5	1.3	0.8–2.3
Current medical insurance
No (ref)	1.0		1.0
Yes	1.9	1.3–2.8	1.2	0.7–2.1
Regular source of medical care
No (ref)	1.0		1.0
Yes	2.6	1.3–5.4	3.1	1.4–7.1
Visited healthcare provider in last 12 months
No (ref)	1.0		1.0
Yes	4.7	2.4–9.2	4.4	2.2–9.0
Provider recommendation ever
Pap smear: No (ref)	1.0		1.0
Pap smear: Yes	3.7	2.4–5.8	5.9	3.3–10.6
Mammogram: No (ref)	1.0		1.0
Mammogram: Yes	4.6	3.0–7.1	6.6	3.7–12.0
Neither (ref)	1.0		1.0
Mammogram only	1.6	0.6–4.6	2.8	0.5–15.6
Pap smear only	1.0	0.4–2.3	2.0	0.8–5.3
Both	4.9	3.0–7.9	8.2	4.3–15.7

Sample size includes only women who have not had a hysterectomy.

Using the results of the unadjusted logistic regression modeling and in consideration of possible confounding, the final models were constructed to include the following variables: country of birth, acculturation level, current medical insurance, visited healthcare provider in the last 12 months for any reason, and ever having had a provider recommendation for breast or cervical cancer screening. Acculturation was included in the multivariate model as the predisposing variable that could best assess differences between Mexican-born and U.S.-born Hispanics regardless of place of birth. Visiting healthcare provider in the last 12 months for any reason and provider recommendation ever were the enabling variables that had the strongest measure of effect in the unadjusted analyses. Current medical insurance was included to account for access to care among the study participants. Thus, the model includes a natural flow from current medical insurance affecting visitation to a healthcare provider in the last 12 months affecting if the provider ever recommends a mammogram or Pap smear.

Results of the multivariate adjusted models are presented in Table 4 as AORs. The results again demonstrated the strength of enabling variables in predicting screening use. Women who reported having visited a healthcare provider in the last 12 months were more likely to report a mammogram within 1 year and Pap smear within 3 years (AOR 3.8, 95% CI 1.7–8.1 and AOR 5.5, 95% CI 2.2–13.7, respectively). Women who received a doctor recommendation for both mammography and Pap smear were more likely to report a mammogram within 1 year and Pap smear within 3 years compared to those who received no recommendation (AOR 5.1, 95% CI 3.0–8.9 and AOR 9.7, 95% CI 4.6–20.7). Having current medical insurance is a statistically significant predictor of mammography within 1 year but not Pap smear within 3 years. Acculturation level and country of birth results were not statistically significant for all screening outcomes.

Table 4.

Mammography and Pap Smear Screening: Adjusted Associations with Predisposing and Enabling Variables Among a Random Sample of Hispanic Women in Yuma County, Arizona

	Mammogram within 1 year (n = 452)		Pap smear within 3 years (n = 271)^a
	AOR ^b	95% CI	AOR ^b	95% CI
Predisposing factors
Country of birth
Mexico (ref)	1.0		1.0
United States	0.5	0.2–1.2	0.4	0.1–1.3
Acculturation
Low (ref)	1.0		1.0
Bicultural or high	1.0	0.5–2.3	2.0	0.6–6.4
Enabling factors
Current medical insurance
No (ref)	1.0		1.0
Yes	1.6	1.0–2.6	1.0	0.5–2.1
Visited healthcare provider in last 12 months
No (ref)	1.0		1.0
Yes	3.8	1.7–8.1	5.5	2.2–13.7
Provider recommendation ever
Neither (ref)	1.0		1.0
Mammogram only	1.7	0.5–5.5	5.7	0.5–65.1
Pap smear only	1.0	0.4–2.5	2.4	0.7–8.1
Both	5.1	3.0–8.9	9.7	4.6–20.7

Sample size includes only women who have not had a hysterectomy.

Adjusted ORs (AOR) mutually adjusted for all variables in table.

Discussion

Breast and cervical cancer screening practices for Hispanics across the United States have been described by various authors. Coughlin and Uhler¹⁶ identified age, race, educational attainment, number of children, health insurance coverage, and physician visit in the last year as predictors of recent mammography and cervical screening use. Lack of insurance, low income, low educational attainment, foreign birth, and low acculturation were all associated with risk of nonuse of cervical cancer screening.^17,18

In our study, enabling characteristics, especially prior clinician recommendation for both mammography and Pap smear and having reported a visit to a clinician in the last year, were significant predictors of older Hispanic women who live near the U.S.-Mexico border reporting mammography and Pap smear screening. The magnitude of the association was strongest for clinician recommendation of both mammography and Pap smear for mammogram within the past year and Pap smear within the past 3 years, (AOR 5.1, 95% CI 3.0–8.9 and AOR 9.7, 95% CI 4.6–20.7). It was also noteworthy that doctor visit within the last year was strongly associated with receiving breast and cervical cancer screening, suggesting that clinicians need to be taking full advantage of medical visits by medically underserved women to inform them of other preventive services.

The importance of enabling factors on breast and cervical cancer screening use has been documented in other studies with Hispanic women. In particular, a physician's recommendation for a mammogram or Pap smear is among the most important determinants of Hispanic women's use of breast and cervical cancer screening.^{16,19

–26} Having current medical insurance and a regular source of medical care are also important predictors of breast and cervical cancer screening use.^23,27
–29 However, our community-based findings of breast and cervical cancer screening use among an older Hispanic female U.S.-Mexico border population are unique to the literature.

We describe a substantial amount of heterogeneity in health risks and cancer screening practices among women living in border communities. Age, acculturation level, education level, marital status, length of residence in the United States, monthly income, current medical insurance, regular source of medical care, self-rated health, and tobacco use ever differ significantly between Mexican-born Hispanic and U.S.-born Hispanic women. Interestingly, the two variables that were the strongest predictors of screening use, visiting clinician in the past year and clinician recommendation for screening, did not significantly differ between Mexican-born and U.S.-born Hispanic women. A large majority of the study sample, 86% of Mexican-born Hispanics and 80% of U.S.-born Hispanics, had visited a healthcare provider in the past year. This is likely due to cross-border use of services. Indeed, although only 47% of Mexican-born Hispanics and 75% of U.S.-born Hispanics stated that they had current medical insurance, 93% and 84%, respectively, reported a regular source of medical care. Previous research also has shown that residents of border communities in many ways have more in common with each other than with their respective countries.¹⁰ Thus, having similar access to clinician services also exposes both Mexican-born and U.S.-born Hispanic women to similar opportunities of receiving clinician recommendations for screening.

Although human papillomavirus (HPV) vaccines appear efficacious for prevention against the acquisition of HPV16 and HPV18, the viral types associated with approximately 70% of cervical cancer, these vaccines will not benefit women in this older age group regardless of vaccine affordability issues.³⁰ Thus, increasing the cancer screening coverage among underserved and underscreened populations must remain a priority. By increasing screening participation among postreproductive-age women through expansion of affordable health services and insurance, the burden of cancer can be reduced in this generation of women through early detection. This strategy may also provide access to daughters and granddaughters who may benefit from vaccination and screening strategies.

There are important policy considerations and practical lessons that can be drawn from this study. Foremost is the observation that opportunistic and combined breast and cervical cancer screening messages are critical among poor, older women with inadequate healthcare coverage. Such bundling of cancer screening may be important and could include other interventions.³¹ The ACS has promoted the concept that the occasion of a general health examination provides a good opportunity to address screening examinations and counseling that could lead to early cancer detection.³² Packaging of screening services may be critical to underserved populations. Furthermore, a recent study suggested that continued Pap smear screening should be considered for women ≥65years among high-risk minority racial/ethnic groups.⁵

There are some limitations to this study. First, because the questionnaires were interviewer administered, social desirability may have biased some responses. Respondents may have answered key questions in an effort to please the interviewer or to be seen as compliant.³³ However, the use of trained promotoras, who followed a consistent script to administer the questionnaires, standardized the data collection procedures and ensured consistent administration of the questionnaires.

Another limitation is self-reported screening use, which can limit the reliability of the outcome variables. By obtaining a medical release to review the last mammography and cervical cytology results from study participants, however, we were able to review and confirm self-reported mammogram in 283 of the 365 women (78%) and self-reported Pap smear in 232 of the 241 women (96%). In addition, previous studies have shown that self-reports of both mammography and Pap smears are valid measures to assess breast and cervical cancer screening history, although greater accuracy for self-reports of mammography than for self-reports of Pap smears has been observed.^34

–40 One study showed that overreporting of mammography screening was highest among African American women and lowest among Hispanic women.³⁵

A related limitation is the possibility of recall bias of key variables. A key item that could be susceptible to recall bias is provider recommendation of breast or cervical cancer screening. However, it is unlikely that there was systematic recall bias in one direction for the self-report of screening behaviors or their predictors. Another limitation is the cross-sectional nature of the study, which limits the interpretation of the results, as the temporal relationship between variables associated with screening outcomes cannot be fully determined. However, where possible, items in the questionnaire were included to attempt to delineate a time sequence.

To address disparities in cancer screening behaviors and burden of disease, public programs promoting breast and cervical cancer prevention in Hispanic women must be tailored to target immigrant and non-English-speaking women. Language and culturally appropriate outreach materials that promote screening are important unmet needs in border populations. We found that opportunistic screening during routine medical visits and after receiving recommendations for mammography and Pap smear screening may increase breast and cervical cancer screening coverage and reduce the burden of these two cancers in U.S.-Mexico Hispanic border populations.

Yuma Project Technical Team

Kenneth Hatch, Susan Vanzzini, Bonnie Monaco, Mario Barragan, and Cora Tadeo, Department of Obstetrics & Gynecology, University of Arizona; Anna Giuliano, Martha Abrahamson, and Linda Vaught, Arizona Cancer Center; Jill Gurnsey de Zapien, College of Public Health, University of Arizona; Jennifer Hunter and Mike Lebowitz, Southwest Border Prevention Center; John Davis, Julie Weins, and Terry Fernandez, Department of Pathology, University of Arizona; Arturo Gonzalez and Scott Carvajal, Mexican American Studies and Research Center, University of Arizona; Catalina Denman and Elena Mendez de Galaz, Colegio de Sonora, Hermosillo, Sonora, Mexico; Amanda Aguirre, Tuly Medina, Veronica Pena, and the Promotora staff of the Western Arizona Area Health Education Center, Yuma.

Footnotes

Acknowledgments

This work was funded by the Center for Medicaid and Medicare Services under contract number 25-P-91062, with additional support provided by the Centers for Disease Control and Prevention Research Center-Southwest Center for Community Health Promotion. We gratefully acknowledge the enthusiastic participation of the women who took part in the study survey. We also recognize the invaluable contributions of the lay community health workers (promotoras) who recruited the survey participants and conducted the interviews. We are indebted to the community of women's healthcare practitioners in Yuma County who provided access to their patients and their clinics for the purposes of this study. We are also indebted to Dr. Richard Bragg at the Centers for Medicare & Medicaid Services (CMS), who provided broad guidance and input on this project. This article was completed as part of dissertation work for the Epidemiology Graduate Program at the Mel and Enid Zuckerman College of Public Health.

Disclosure Statement

The authors declare that no competing financial interests exist.

References

National Center for Health Statistics. Health, United States 2006 with chartbook on trends in the health of Americans. Hyattsville, MD: U.S. Department of Health and Human Services, 2006.

American Cancer Society. Cancer facts & figures for Hispanics/Latinos 2009–2011. Atlanta: American Cancer Society, 2009.

Hedeen

, White

. Breast cancer size and stage in Hispanic American women, by birthplace: 1992–1995. Am J Public Health, 2001; 91:122–125.

Lantz

, Mujahld

, Schwartz

et al. The influence of race, ethnicity, and individual socioeconomic factors on breast cancer stage and diagnosis. Am J Public Health, 2006; 96:2173–2178.

Barnholtz-Sloan

, Patel

, Rollison

, Kortepeter

, MacKinnon

, Giuliano

. Incidence trends of invasive cervical cancer in the United States by combined race and ethnicity. Cancer Causes Control, 2009; 20:1129–1138.

Seeff

, McKenna

. Cervical cancer mortality among foreign-born women living in the United States, 1985 to 1996. Cancer Detect Prev, 2003; 27:203–208.

Coughlin

, Richards

, Nasseri

et al. Cervical cancer incidence in the United States in the US-Mexico border region, 1998–2003. Cancer, 2008; 113,Suppl 10:2964–2973.

Wallace

, Hunter

, Papenfuss

, de Zapien

, Denman

, Giuliano

. Pap smear screening among women ≥40 years residing at the United States-Mexico border. Health Care Women Int., 2007; 28:799–816.

Fernandez

, Morales

. Language and use of cancer screening services among border and non-border Hispanic Texas women. Ethnicity Health, 2007; 12:245–263.

10.

Hunter

, De Zapien

, Denman

et al. Healthcare access and utilization among women 40 and older at the U.S.-Mexico border: Predictors of a routine check-up. J Community Health, 2003; 28:317–333.

11.

U.S.-Mexico Border Health Commission. Healthy border 2010. An agenda improving health on the United States-Mexico border. 2003.

12.

Gelberg

, Andersen

, Leake

. The behavioral model for vulnerable populations: Application to medical care use and outcomes for homeless people. Health Serv Res, 2000; 34:1273–1302.

13.

Andersen

. Revisiting the behavioral model and access to medical care: Does it matter? J Health Soc Behav, 1995; 36:1–10.

14.

Balcazar

, Castro

, Krull

. Cancer risk reduction in Mexican American women: The role of acculturation, education, and health risk factors. Health Educ Behav, 1995; 22:61–84.

15.

Smith

, Cokkinides

, Brawley

. 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; 59:27–41.

16.

Coughlin

, Uhler

. Breast and cervical cancer screening practices among Hispanic women in the United States and Puerto Rico, 1998–1999. Prev Med., 2002; 34:242–251.

17.

Hewitt

, Devesa

, Breen

. Papanicolaou test use among reproductive-age women at high risk for cervical cancer: Analyses of the 1995 National Survey of Family Growth. Am J Public Health, 2002; 92:666–669.

18.

Shah

, Zhu

, Wu

, Potter

. Hispanic acculturation and utilization of cervical cancer screening in the United States. Prev Med, 2006; 42:146–149.

19.

Mandelblatt

, Gold

, O'Malley

et al. Breast and cervix cancer screening among multiethnic women: Role of age, health, and source of care. Prev Med, 1999; 28:418–425.

20.

Magai

, Consedine

, Conway

, Neugut

, Culver

. Diversity matters: Unique populations of women and breast cancer screening. Cancer, 2004; 100:2300–2307.

21.

Borrayo

, Thomas

, Lawsin

. Cervical cancer screening among Latinas: The importance of referral and participation in parallel cancer screening behaviors. Women Health, 2004; 39:13–29.

22.

Sung

, Alema-Mensah

, Blumenthal

. Inner-city African American women who failed to receive cancer screening following a culturally appropriate intervention: The role of health insurance. Cancer Detect Prev, 2002; 26:28–32.

23.

Aldridge

, Daniels

, Jukic

. Mammograms and healthcare access among US Hispanic and non-Hispanic women 40 years and older. Fam Community Health, 2006; 29:80–88.

24.

Tejeda

, Thompson

, Coronado

, Martin

, Heagerty

. Predisposing and enabling factors associated with mammography use among Hispanic and non-Hispanic white women living in a rural area. J Rural Health, 2009; 25:85–92.

25.

Bazargan

, Bazargan

, Farooq

, Baker

. Correlates of cervical cancer screening among underserved Hispanic and African-American women. Prev Med, 2004; 39:465–473.

26.

Zapka

, Harris

, Hosmer

, Costanza

, Mas

, Barth

. Effect of a community health center intervention on breast cancer screening among Hispanic American women. Health Serv Res, 1993; 28:223–235.

27.

Abraido-Lanza

, Chao

, Gammon

. Breast and cervical cancer screening among Latinas and non-Latina whites. Am J Public Health, 2004; 94:1393–1398.

28.

Coughlin

, Uhler

, Richards

, Wilson

. Breast and cervical cancer screening practices among Hispanic and non-Hispanic women residing near the United States-Mexico border, 1999–2000. Fam Community Health, 2003; 26:130–139.

29.

, Black

, Markides

. Prevalence and associated factors of cancer screening: Why are so many older Mexican American women never screened? Prev Med, 2001; 33:268–273.

30.

Saslow

, Castle

, Cox

et al. American Cancer Society guidelines for human papillomavirus (HPV) vaccine use to prevent cervical cancer and its precursors. CA Cancer J Clin, 2007; 57:7–28.

31.

Kim

, Salomon

, Weinstein

, Goldie

. Packaging health services when resources are limited: The example of a cervical cancer screening visit. PLoS Med, 2006; 3:e434.

32.

Smith

, Cokkinides

, Brawley

. Cancer screening in the United States, 2008: A review of current American Cancer Society guidelines and cancer screening issues. CA Cancer J Clin, 2008; 58:161–179.

33.

Johnson

, O'Rourke

, Burris

, Warnecke

. An investigation of the effects of social desirability on the validity of self-reports of cancer screening behaviors. Med Care, 2005; 43:565–573.

34.

Norman

, Localio

, Zhou

et al. Validation of self-reported screening mammography histories among women with and without breast cancer. Am J Epidemiol, 2003; 158:264–271.

35.

Cronin

, Miglioretti

, Krapcho

et al. Bias associated with self-report of prior screening mammography. Cancer Epidemiol Biomarkers Prev, 2009; 18:1699–1705.

36.

Warnecke

, Sudman

, Johnson

et al. Cognitive aspects of recalling and reporting health-related events: Papanicolaou smears, clinical breast examinations, and mammograms. Am J Epidemiol, 1997; 146:982–992.

37.

Gordon

, Hiatt

, Lampert

. Concordance of self-reported data and medical record audit for six cancer screening procedures. J Natl Cancer Inst, 1993; 85:566–570.

38.

Caplan

, McQueen

, Qualters

, Leff

, Garrett

, Calonge

. Validity of women's self-reports of cancer screening test utilization in a managed care population. Cancer Epidemiol Biomarkers Prev, 2003; 12:1182–1187.

39.

Paskett

, Tatum

, Mack

, Hoen

, Case

, Velez

. Validation of self-reported breast and cervical cancer screening tests among low-income minority women. Cancer Epidemiol Biomarkers Prev, 1996; 5:721–726.

40.

Suarez

, Goldman

, Weiss

. Validity of Pap smear and mammogram self-reports in a low-income Hispanic population. Am J Prev Med, 1995; 1:94–98.