• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of amjphAmerican Journal of Public Health Web SiteAmerican Public Health Association Web SiteSubmissionsSubscriptionsAbout Us
Am J Public Health. 2005 December; 95(12): 2225–2230.
PMCID: PMC1449511

Breast Cancer Treatment Experiences of Latinas in Los Angeles County

Abstract

Objective. We examined breast cancer treatment experiences of and outcomes for Latinas in Los Angeles County.

Methods. We conducted a population-based survey of women who were diagnosed with breast cancer between December 2001 and November 2002 (n=910) to evaluate the types of treatments received, communication with clinicians, and satisfaction.

Results. About two thirds were non-Latina White, 18.8% were African American, and 18.9% were Latina (with 11.0% preferring English and 7.9% preferring Spanish). The rest indicated other ethnic groups. Latinas who preferred Spanish were more likely to experience a delay of 3 months or more from diagnosis to surgical treatment (36.4% vs 9.1% for non-Latina Whites, 18.6% for African Americans, and 12.7%, for other Latinas, P<.001). African Americans and Latinas who preferred Spanish had very low rates of reconstruction (13.8% and 9.2%, respectively, compared with 42.1% for Whites and 34.5% for Latinas who preferred English, P=.009). Latinas who preferred Spanish had the highest odds ratio for low satisfaction.

Conclusion. Latinas who preferred Spanish received different treatments and perceived a different treatment experience than did other cultural groups.

Breast cancer is the most common cause of cancer and cancer-related mortality among Latinas in the United States. In 2003, about 11000 new cases of breast cancer were diagnosed in Latinas—a higher incidence than for the next 4 cancer sites combined.1 Population-based studies have shown that Latinas present with more advanced disease and with more adverse prognostic indicators than do non-Latina White women.2,3 A recent analysis using data from the Surveillance, Epidemiology, and End Results (SEER) Program showed that Latinas had a higher risk of mortality after a breast cancer diagnosis than did non-Latina White women.4 Further, Latinas were more likely to receive a first course of treatment that did not meet 2000 National Comprehensive Cancer Network standards.4

Ethnic differences in breast cancer morbidity and mortality are likely the result of a complex interplay between clinical, cultural, and socioeconomic factors.5,6 First, Latina women present with later-stage disease that may limit treatment options.3,7,8 In addition, powerful cultural factors shape the health care treatment experiences of many Latinas, who have diverse levels of acculturation as expressed by language, family origin, immigration experience, and social class. Patients with low acculturation are more likely to prefer to communicate in Spanish. Studies have shown that language and literacy are major barriers to cancer screening and treatment opportunities.9

Communication between patients and clinicians may also be influenced by a number of other cultural factors that are relevant to the Hispanic community and may affect patient preferences with regard to providers or their perceptions of quality of care.10,11 Finally, financial and nonfinancial dimensions of access are highly correlated with acculturation and strongly limit opportunities to obtain quality treatment.12 Although there is a growing literature on African American versus White differences in breast cancer treatment, there is very little information to date on Latinas’ experiences with such treatment, especially from population-based studies.13

To explore Latinas’ experiences with and perspectives on breast cancer treatment and communication, we conducted a population-based study to examine the following questions: (1) Were there differences in the distribution of specific therapies (mastectomy vs breast-conserving surgery [BCS], breast reconstruction, and radiation post-BCS) across ethnic groups? (2) What patient–surgeon communication factors were associated with ethnic differences in receipt of therapies? (3) Were ethnic differences in treatment associated with socioeconomic and access factors? (4) Were there ethnic differences in satisfaction with the decision-making process regarding surgery?

METHODS

Database and Sampling

All women with ductal carcinoma in situ (DCIS) and a random sample of those with invasive breast cancer (oversampling African Americans) reported to the SEER registry of Los Angeles County were prospectively accrued each month from December 2001 through November 2002 (n = 1311). Eligible subjects underwent a definitive surgical procedure, resided in the catchment area of the SEER registry, and were able to complete a questionnaire in English or Spanish. All Asian women and all US-born women younger than 50 years diagnosed with invasive disease during our study period were excluded because they were already being enrolled in other studies. Women with a diagnosis of lobular carcinoma in situ were excluded as well because the natural history of and recommended treatment for this disease are different than they are for DCIS. Ninety percent of all accrued cases were eligible for the study (n = 1170), and the survey was completed by 77.8% of eligible patients (N = 910).

Data Collection and Management

Physicians were notified of our intent to contact patients; notification was followed by an introductory letter and a telephone call to potential subjects to assess eligibility. A questionnaire and a $10 gift certificate were mailed to eligible women who agreed to participate and to those who could not be reached by phone. The Dillman survey method, which involves follow-up reminders and subsequent mailings to nonrespondents at periodic intervals, was employed to enhance response.14

All survey materials were translated into Spanish by experienced native speakers of Mexican descent. Speakers of Central and South American descent performed a cross-cultural check for appropriateness. Back-translation was performed in iterations, and differences in versions were evaluated and reconciled to ensure that the original meaning had been retained. Subsequently, the Los Angeles site coordinator (a native speaker of Mexican descent) screened all women accrued into the study and ascertained whether potential respondents preferred an English or a Spanish version of the questionnaire. This process was repeated during telephone recontacts with those who failed to return a questionnaire and with those who returned incomplete questionnaires to determine the appropriate language for a telephone interview or clarification of missing data. A total of 72 of 104 patients (69.2%) who were sent a Spanish questionnaire completed and returned it.

SEER registry clinical data submitted by cancer-reporting facilities in Los Angeles County were merged with survey data for 98.2% of the cases.

Measures

Dependent variables included measures of surgical and adjuvant radiation therapy. The surgical treatment received (unilateral or bilateral mastectomy vs BCS) was ascertained by self-report and the SEER data, which yielded the same surgical procedure for 96.4% of the patients in our sample (a rate consistent with other studies).15 Initiation of radiation therapy was ascertained by patient self-report or from SEER clinical data if missing on the survey (n = 18). Receipt of breast reconstruction (dichotomous variable) was based on self-report. We also created a variable indicating delay, defined as a duration of 3 months or longer between diagnosis and definitive surgical treatment.

Another dependent variable, satisfaction with the process by which the surgical decision was made, was measured by means of a scale comprising 4 attitudinal items from the survey identified from a larger pool of items through data reduction. A 5-point Likert scale (ranging from strongly disagree to strongly agree) was used to measure the level of agreement with the following 4 items addressing the decision on what type of surgery to have: (a) I wish I would have given more consideration to other options; (b) I would have liked to have had more information; (c) I would have liked to have been more active in the decision-making process; (d) I did not have as much to say as I would have liked. The Cronbach α for this scale was .91 (.91, .89, .91 for non-Latina White, African American, and Latina patients, respectively), with the scale score representing the average response to the 4 items.

Independent variables included clinical factors. Summary cancer stage was classified using the American Joint Committee on Cancer TNM (Tumor, Node, Metastasis) Staging System for breast cancer.16 A count of medical comorbidities was based on a list of 6 chronic conditions selected from the 2001 National Health Interview Survey.17 Sociodemographic variables included age (continuous), education (4 categories, see Table 1 [triangle]), marital status (currently married/partnered or not), household income (5 categories), and insurance status (none, private/Medicare, Medicaid). Ethnicity was determined by self-report. Categories for analysis included White, African American, and Latina (self-identified). Latina women were further categorized according to whether they completed the survey in English (Latina-English) or Spanish (Latina-Spanish).

TABLE 1
Study Sample Characteristics (%), by Ethnicity

Analysis

Fourteen patients who reported an ethnic group other than White, African American, or Latina or were missing data were excluded from the analyses. We first analyzed sample population characteristics by ethnic groups: We examined the use of mastectomy and treatment delay by ethnic groups in separate logistic regression models with control for clinical and sociodemographic factors. We then examined correlates of the use of breast reconstruction after mastectomy and the use of radiation after BCS. We next evaluated patient perceptions of discussions with their surgeons about surgical and radiation treatment by ethnic group. Finally, we examined ethnic differences in perceived satisfaction with the decisionmaking process. The distribution of scale scores was skewed, with the majority of respondents reporting fairly high levels of satisfaction. Thus, we dichotomized the outcome variable and conducted a logistic regression analysis to evaluate the independent association of satisfaction with ethnicity, with control for clinical and sociodemographic characteristics. Low satisfaction was defined as a scale score indicating that the average answer to scale items was neutral, agree, or strongly agree. Other cut points in the logistic approach and other approaches to modeling, including ordinary least squares with log transformation of the dependent variable, yielded consistent results. We evaluated all models for second-order interactions and used sampling weights to account for differential selection and nonresponse.

RESULTS

Sample Characteristics

About two thirds (60.8%) of respondents were non-Latina White, 18.8% were African American, 18.9% self-designated as Latina (11.0% Latina-English and 7.9% Latina-Spanish), and 1.5% indicated other ethnic groups. Table 1 [triangle] shows the weighted percentage distribution of the characteristics of the sample population by ethnic group. Ethnic differences were significant for all categories (P < .001). Compared with all other ethnic groups, Latina-Spanish patients were younger, had much lower levels of education, had much lower family incomes, and were much more likely to be uninsured or have Medicaid.

Table 1 [triangle] also shows that there was modest ethnic variation in the prevalence of medical comorbidities, with African Americans having the highest prevalence. Latina-Spanish patients presented with a later stage at diagnosis than those from other ethnic groups (27.1% with stage 3 disease compared with about 10% for all other ethnic groups), and more of them had a mastectomy than other ethnic groups. However, the proportion of Latina-Spanish patients who received reconstruction after mastectomy was very low. Finally, there was modest ethnic variation in the receipt of radiation after BCS.

Treatment

Overall, 32.2% of patients received mastectomy, and after control for clinical and sociodemographic factors, this percentage did not vary significantly by ethnic group. However, Latina-Spanish patients were much more likely to experience a delay between diagnosis and surgery of 3 months or more (36.4% vs 9.1% for non-Latina Whites, 18.6% for African Americans, and 12.7% for Latina-English patients, P < .001). Table 2 [triangle] shows 2 models with adjusted odds ratios for treatment delay by ethnic group. Model 1 controlled for clinical and demographic variables. Model 2 built on Model 1 by adding education, income, and insurance status variables, primarily to see what happened to the ethnicity variable after control for these other factors. The results show that treatment delay for Latina-Spanish patients substantially attenuated to insignificance with the addition of these variables in Model 2, suggesting that socioeconomic status—including insurance status—explains ethnic difference in delay between diagnosis and surgery in this sample. Insurance status had a strong and significant effect on treatment delay (none/Medicaid vs private/Medicare, adjusted odds ratio [AOR] = 2.3, 95% confidence interval [CI] = 1.1, 5.1).

TABLE 2
Adjusted Odds Ratios With Confidence Intervals for Surgical Treatment Delay Exceeding 3 Months

The percentage of patients who received breast reconstruction after mastectomy varied by ethnic group, adjusted for clinical and sociodemographic factors: non-Latina White patients had the highest percentage (42.1%) followed by Latina-English patients (34.5%). African American and Latina-Spanish patients had very low rates of reconstruction (13.8% and 9.2%, respectively, Wald test 11.5, P = .009). Latina-Spanish patients were less likely to report that they discussed reconstruction with their surgeons (42.2% vs 70.0% for non-Latina Whites, 55.4% for African Americans, and 56.3% for Latina-English patients, P = .014). Table 3 [triangle] shows AORS for no receipt of reconstruction after mastectomy. In Model 1, we controlled for clinical and demographic factors. In Model 2, education, income, and insurance status variables were added. Table 3 [triangle] shows that with the addition of the socioeconomic and insurance variables, the odds of not receiving reconstruction greatly attenuated for Latinas, but not for African Americans. Insurance status was a significant factor in Model 2 (none/Medicaid vs private/Medicare, AOR = 5.9, 95% CI = 1.1, 20.0).

TABLE 3
Adjusted Odds Ratios With Confidence Intervals for No Receipt of Reconstruction After Mastectomy

Of patients who received BCS, 96% of those with invasive breast cancer reported that they received a recommendation for radiation, compared with 72.2% of those with DCIS (P < .001). About 85% of women with invasive disease reported receiving radiation, compared with 60.9% for DCIS (P < .001). There was no consistent association between income or insurance and radiation after BCS. Figure 1 [triangle] shows the percentages of patients who reported a surgeon recommendation and who received radiation by stage and ethnicity. There was a trend toward less use of radiation for Latina women with DCIS than for their non-Latina White counterparts. Latina-Spanish patients were also less likely to recall a recommendation for radiation than women in other ethnic groups. However, there was little ethnic variation in recommendation or receipt of radiation for women with invasive disease.

FIGURE 1.
Surgeon recommendation of radiation and receipt of radiation after breast-conserving surgery, by ethnicity and tumor behavior.

Satisfaction

After we controlled for age, education, and stage, we calculated odds ratios for low satisfaction with the treatment decision process by ethnic group. Compared with non-Latina Whites (the reference group), Latina-Spanish patients had the highest odds ratio (OR = 3.6, 95% CI = 2.9, 6.9), followed by African American patients (OR = 2.2, 95% CI = 1.7, 3.9) and Latina-English patients (OR = 1.3, 95% CI = 1.0, 1.9, Wald test for ethnic differences 13.2, P < .001). Education was inversely associated with low satisfaction (AORs = 1.0 for college or more, 0.9 for some college, 2.0 for high school graduate, 2.2 for less than high school graduate, Wald test 8.0, P = .018). However, patient age and stage of disease were not associated with low satisfaction.

DISCUSSION

We analyzed population-based data linking tumor registry and mailed survey data to investigate ethnic differences in treatment experiences pertaining to surgery and radiation therapy for women diagnosed with breast cancer in Los Angeles County in 2002. We found substantial and significant differences in experience by ethnic group. Compared with other ethnic groups, Latinas who preferred to communicate in Spanish experienced substantial delay in definitive surgery. Ethnic differences in delay decreased markedly when we adjusted for income and insurance status. Further, Latinas who preferred Spanish reported the lowest receipt of breast reconstruction after mastectomy, which was also associated with insurance status. The findings for receipt of radiation after BCS were less consistent. After adjusting for clinical and sociodemographic factors, Latina-Spanish patients reported somewhat lower receipt of radiation after BCS. These differences were consistent with patient recall of discussions and recommendations by their surgeons. Finally, although patient satisfaction was generally high, Latinas who preferred Spanish reported lower satisfaction than did women in other ethnic groups.

Factors Influencing the Treatment Experience

Our findings suggest that a number of factors may shape the treatment experience of our Los Angeles sample of Latinas with breast cancer. Spanish-speaking Latinas were younger and had much less education than women in other ethnic groups. These sociodemographic factors have been correlated with treatment receipt and experience in other studies.18 Latina-Spanish patients also presented with a later stage at diagnosis, which may limit treatment choices. However, ethnic disparities were observed even after we controlled for these factors.

Financial and nonfinancial barriers to health care access likely play important roles in ethnic differences in breast cancer treatment.12 Indeed, we found that income and insurance status were important confounding factors in the association between ethnicity, delayed surgical treatment, and reconstruction after mastectomy. Financial access is an important factor to consider in investigating ethnicity and cancer treatment. Similarly, access to quality health care is another important dimension because many Latinas live in neighborhoods that are far from comprehensive cancer centers or they may have limited access to quality cancer care for other reasons. Additional dimensions of access that may play a role include patient accommodation (evening appointments and translation services), availability of transportation, and patient acceptability of providers and the health care environment.19 In particular, limited communication skills and understanding of Hispanic culture by providers may also create barriers to care. These factors can lead to patient dissatisfaction with the treatment that was received and the way that medical decisions were made.

We can only speculate about other cultural and access factors that may have helped shape the experience of Spanish-preferring Latinas in our study. Limited facility with English is a powerful barrier to communication and quality medical care in the United States.20 Thus, some of our findings may be explained primarily by language barriers, as studies have shown that language and literacy are major barriers to cancer screening and treatment opportunities.9 Language and literacy may be particularly challenging in treating breast cancer because of the complexity of the information and decisionmaking process.

Language is also a powerful indicator of patient acculturation. There is a rich literature about the important features of Hispanic culture in the United States and how these cultural factors may influence patient care.10,11 Communication between patients and clinicians may be influenced by a number of cultural factors such as “personalismo” (the emphasis on personal manner in communication and relationships), “simpatia” (the tendency to appraise the quality of relationships and communication on the basis of friendliness and warmth), or power distance (deference to people in positions of perceived power). These values may limit involvement in decisionmaking details, resulting in poor knowledge about treatment options and trade-offs and, ultimately, lower patient satisfaction.

Limitations

Several limitations merit comment. Although the study was population based, the exclusion of Asian women and US-born women aged younger than 50 limits general-izability to these groups. The fact that the percentage of Latina women, particularly those who preferred Spanish, was modest may have limited power for some analyses. Furthermore, somewhat lower response rates for African Americans and Spanish-preferring Latinas compared with Whites may have introduced bias. We did not have robust measures of acculturation, such as patient origin or immigration experience, and thus our evaluations of some of the underlying factors driving the differences we observed were limited.

The study was necessarily retrospective. Patients’ recall of their encounters with clinicians may vary because of the passage of time or be influenced by their posttreatment experiences. However, the average time from treatment to completion of the questionnaire was 7 months (range 1 to 14 months), and there was no association between the time from diagnosis date to questionnaire completion date and the findings.

Conclusions

Our results suggest that clinicians and other providers face special challenges when caring for Latinas with breast cancer, especially those with low acculturation. A cancer diagnosis is a stressful experience for all patients. They and their families must make many complicated treatment decisions with major consequences over a short period of time. Most of these decisions are made with clinicians and staff they have met for the first time under difficult circumstances. A major challenge to those who work with patients who have breast cancer is that many Latina women with low acculturation may be engaging the health care system for the first time with little or no health insurance and very limited financial resources. Although these daunting access issues are common to patients from other vulnerable populations, providers face additional challenges related to language and culture. More research is needed to identify the underlying factors driving disparities in treatment for Latinas, communication, and outcomes in breast cancer and to inform initiatives to improve the quality of care.

Acknowledgments

This work was funded by a grant from the National Cancer Institute (RO1 CA8837–A1) to the University of Michigan and was also funded in part by the National Cancer Institute (contract nos. N01–PC–35139 and NO1–PC–65064). The collection of cancer incidence data was supported by the California Department of Health Services as part of the statewide cancer-reporting program mandated by California Health and Safety Code §103885.

Note. The ideas and opinions expressed herein are those of the authors, and no endorsement by the California Department of Health Services is intended or should be inferred.

Human Participant Protection
The study protocol was approved by the institutional review boards of the University of Michigan, the University of Southern California, and Wayne State University.

Notes

Peer Reviewed

Contributors
S. J. Katz, P. M. Lantz, N. K. Janz, and A. Fagerlin were involved in the project planning, implementation, analyses, and writing of this article. D. Deapen, L. Liu, and Y. Paredes were involved in the project planning, implementation, and writing of this article.

References

1. Cancer Facts and Figures for Hispanics/Latinos, 2003–2005. Atlanta, Ga: American Cancer Society; 2003. Available at: http://www.cancer.org/downloads/STT/CAFF2003HispPWSecured.pdf. Accessed October 5, 2005.
2. Shavers VL, Harlan LC, Stevens JL. Racial/ethnic variations in clinical presentation, treatment and survival among breast cancer patients under age 35. Cancer. 2003;97:134–147. [PubMed]
3. Boyer-Chammard A, Taylor TH, Anton-Culver H. Survival differences in breast cancer among racial/ethnic groups: a population-based study. Cancer Detect Prev. 1999;23(6):463–473. [PubMed]
4. Li CI, Malone KE, Daling JR. Differences in breast cancer stage, treatment and survival by race and ethnicity. Arch Intern Med. 2003;163:49–56. [PubMed]
5. Shavers VL, Brown ML. Racial and ethnic disparities in the receipt of cancer treatment. J Natl Cancer Inst. 2002;94(5):334–357. [PubMed]
6. Freeman HP. Race, poverty, and cancer. J Natl Cancer Inst. 1990;83:526–527. [PubMed]
7. Bentley JR, Delfino RJ, Taylor TH. Differences in breast cancer stage at diagnosis between non-Hispanic White and Hispanic populations, San Diego County 1988–1993. Breast Cancer Res Treatment. 1998;50:1–9. [PubMed]
8. Elledge RM, Clark GM. Tumor biologic factors and breast cancer prognosis among White, Hispanic, and Black women in the United States. J Natl Cancer Inst. 1994;86(9):705–712. [PubMed]
9. Seijo R, Gomez H, Freidenberg J. Language as a communication barrier in medical care for Hispanic patients. Hispanic J Behav Sci. 1991;13(4):363–376.
10. Marin G, Van Oss Marin B. Research with Hispanic populations. Applied Social Research Methods Series. 1991;23.
11. Chong N. The Latino Patient: A Cultural Guide for Health Care Providers. Yarmouth, Me: Intercultural Press, Inc.; 2002.
12. Breen N, Kesser LG, Brown ML. Breast cancer control among the underserved—an overview [review]. Breast Cancer Res Treatment. 1996;40(1):105–115. [PubMed]
13. Ramirez AG, Villarreal R, Suarez L, Flores ET. The emerging Hispanic population: a foundation for cancer prevention and control [review]. J Natl Cancer Inst Monographs. 1995;18:1–9. [PubMed]
14. Anema MG, Brown BE. Increasing survey responses using the total design method. J Continuing Education Nurs. 1995;26:109–114. [PubMed]
15. M andandelblatt JS, Berg CD, Meropol NJ, et al. Measuring and predicting surgeons’ practice styles for breast cancer treatment in older women. Med Care. 2001;39(3):228–242. [PubMed]
16. American Joint Committee on Cancer. Cancer Staging Manual. 5th ed. 1997:172–177.
17. National Center for Health Statistics, Division of Health Interview Statistics. 2001 National Health Interview Survey, public use data release. Available at: ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/2001. Accessed August 20, 2005.
18. Groman R, Ginsburg J. Racial and ethnic disparities in health care: a position paper of the American College of Physicians. Ann Intern Med. 2004;141(3): 226–232. [PubMed]
19. Penchansky R, Thomas JW. The concept of access: definition and relationships to consumer satisfaction. Med Care. 1981;19(2):127–141. [PubMed]
20. Documet PI, Sharma RK. Latinos’ health care access: financial and cultural barriers. J Immigrant Health. 2004;6(1):5–13. [PubMed]

Articles from American Journal of Public Health are provided here courtesy of American Public Health Association
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...