SEARCH

SEARCH BY CITATION

Keywords:

  • breast cancer;
  • patient navigation;
  • health disparities;
  • Latinas;
  • Breast Imaging Reporting and Data System;
  • BI-RADS

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

BACKGROUND.

Breast cancer is the leading cause of cancer-related deaths in Latinas, chiefly because of later diagnosis. The time from screening to diagnosis is critical to optimizing cancer care, yet the efficacy of navigation in reducing it is insufficiently documented. Here, the authors evaluate a culturally sensitive patient navigation program to reduce the time to diagnosis and increase the proportions of women diagnosed within 30 days and 60 days.

METHODS.

The authors analyzed 425 Latinas who had Breast Imaging Reporting and Data System (BI-RADS) radiologic abnormalities categorized as BI-RADS-3, BI-RADS-4, or BI-RADS-5 from July 2008 to January 2011. There were 217 women in the navigated group and 208 women in the control group. Women were navigated by locally trained navigators or were not navigated (data for this group were abstracted from charts). The Kaplan-Meier method, Cox proportional hazards regression, and logistic regression were used to determine differences between groups.

RESULTS.

The time to diagnosis was shorter in the navigated group (mean, 32.5 days vs 44.6 days in the control group; hazard ratio, 1.32; P = .007). Stratified analysis revealed that navigation significantly shortened the time to diagnosis among women who had BI-RADS-3 radiologic abnormalities (mean, 21.3 days vs 63.0 days; hazard ratio, 2.42; P < .001) but not among those who had BI-RADS-4 or BI-RADS-5 radiologic abnormalities (mean, 37.6 days vs 36.9 days; hazard ratio, 0.98; P = .989). Timely diagnosis occurred more frequently among navigated Latinas (within 30 days: 67.3% vs 57.7%; P = .045; within 60 days: 86.2% vs 78.4%; P = .023). This was driven by the BI-RADS-3 strata (within 30 days: 83.6% vs 50%; P < .001; within 60 days: 94.5% vs 67.2%; P < .001). A lack of missed appointments was associated with timely diagnosis.

CONCLUSIONS.

Patient-centered navigation to assist Latina women with abnormal screening mammograms appeared to reduced the time to diagnosis and increase rates of timely diagnosis overall. However, in stratified analyses, only navigated Latinas with an initial BI-RADS-3 screen benefited, probably because of a reduction in missed diagnostic appointments. Cancer 2013. © 2012 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Breast cancer is the most commonly diagnosed cancer among Latinas with an estimated 14,200 diagnosed in 2009.1 Since 1997, incidence has decreased 1.5% per year among non-Latino white (NLW) women but only 0.9% per year among Latinas.1 Approximately 2,200 Latinas died from breast cancer in 2009.2 It is the leading cause of cancer death in Latinas, who are more likely to die of diagnosed breast cancer than NLW women,3 because breast cancer is diagnosed later in Latinas than NLW women. From 2002 to 2006, 55% of Latina breast cancers were diagnosed at the local stage, compared with 63% of NLW breast cancers.4 Additional observations suggest that this is accompanied by larger tumors in Latinas.5, 6 Differences are linked to cultural factors, resulting in delayed screening and follow-up.2, 7 Studies have demonstrated that 90-day diagnostic delays—frequently experienced by Latinos and other minorities—reduce survival rates among patients with cancer.8-10 This is an important cancer health disparity.11

When this disparity is controlled in analyses, Latinas remain approximately 20% less likely to survive breast cancer than NLW women.12 Differences again are associated with cultural and socioeconomic factors that marginalize Latinas and other minorities from cancer care. Language barriers, Breast Imaging Reporting and Data System (BI-RADS) category assigned to the mammogram, insurance status, and other difficulties contribute to suboptimal use of cancer care services.7, 13, 14 This is significant, because Latinos are currently the largest US minority and, by 2030, will constitute an estimated 33% of the nation's population.15, 16 Fueled by psychosocial, linguistic, and other sociocultural barriers, disparities translate into increasingly larger gaps in access to care along the entire cancer care continuum.17, 18

A significant component of these gaps involves the time from abnormal breast screening to diagnosis. Uninsured or underinsured and racial/ethnic minorities often have the longest delays,7, 19, 20 particularly involving ambiguous screening results.21, 22 The literature is limited in Latino cultural focus, consistency in reported outcomes, and small or single-site samples. To address this problem, the federally funded “Redes En Accion: The National Latino Cancer Research Network” tested a culturally tailored patient navigation (PN) intervention model, the Six Cities Patient Navigation Study. On the basis of Harold Freeman's successful navigation model,23 we used trained, bilingual community health workers to assist Latinas in using cancer care services in cities with significant Latino populations (San Francisco, San Diego, New York City, Miami, Houston, and San Antonio, Texas). Study leaders in San Antonio and regional partners in 5 other cities applied findings from previous studies to develop a tailored PN program for Latina women to evaluate its effect on the time from initial abnormal breast screen to the time of diagnosis compared with non-navigated women.24, 25 We hypothesized that navigation would reduce the time from initial abnormal breast screening to diagnosis and would increase rates of timely diagnosis within 30 days or 60 days of an initial abnormal screen compared with standard care.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Study Design and Participants

We used a quasiexperimental design to compare unmatched control participants and intervention participants on the time from abnormal breast screening to diagnosis and the proportions diagnosed within 30 days and 60 days of the initial screen in a multisite study of 480 Latinas (n = 251 navigated; n = 229 non-navigated controls). Data were collected by face-to-face interviews (at baseline) and medical charts (follow-up) among navigated women and medical chart review among non-navigated women in 1 community-based clinic at each site. Clinics were associated with a public health department or an academic medical center. The institutional review boards or equivalent oversight bodies at each of the Six Cities Study sites and each participating community health clinic reviewed and approved this study, including informed consent.

A convenience sampling approach was used to recruit participants. Eligibility criteria targeted 480 self-identified Latinas at community-based health clinics, aged ≥18 years with an abnormal breast screening mammogram resulting in a BI-RADS-3 (“probably benign”), BI-RADS-4 (“suspicious”), or BI-RADS-5 (“highly suggestive of malignancy”) result between January 2008 and January 2011. We excluded women who had a history of cancer treated within the last 5 years on the basis that these women may have had prior experience with the complex cancer treatment system and likely would need less navigation intervention. We also excluded those who had experienced past navigation after an abnormal screening because of the potential for confounding. Pregnant women were not specifically recruited but were not excluded from eligibility.

Controls were chosen by determining eligibility consecutively backward from the study start date. Navigated women were identified from the same clinics by determining eligibility consecutively forward. Individuals in the control group received usual treatment, whereas individuals in the intervention group received usual treatment plus patient navigation. Participants overwhelmingly preferred to speak Spanish and had similar low levels of socioeconomic status, age, employment, marital status, and country of origin.

Navigation

PN was adapted from the Freeman and Rodriguez 4-part case management model.23 Six bilingual Latina patient navigators were used. They all were women ages 25 to 47 years with at least a high school diploma or college degree. In addition, they were trained to coordinate care for those referred for diagnostic evaluation according to our adaptation of the Freeman and Rodriguez model.23 PN generally began 1 week or sooner after the index abnormal screening examination and continued through diagnosis. Navigators emphasized adherence to diagnostic and treatment plans and also influenced patient behavior through effective communication, education, rapport building, and empathy. These techniques were used to assist in overcoming potential barriers, such as lack of transportation and/or child care, inadequate communication with health care providers, insurance coverage issues, fear of test results and other detriments to care. The patient navigators provided culturally sensitive support and guidance to facilitate diagnosis and care; and they also tracked patient demographics and other information, addressing barriers by negotiating, coordinating, and/or providing resources like transportation, support, and advice and serving as an advocate and liaison. PN time was invariant between navigators but varied with individual needs, ranging from 8 minutes to 30 hours and averaging 3.0 ± 5.2 hours (median = 1.1 hour).

Data

The study began in January 2008. Data were collected through a combination of baseline interviews and/or chart abstraction (for navigated women) or retrospective chart abstraction for non-navigated controls. Project coordinators at each site reviewed data abstraction forms and interview records for completeness, accuracy, and internal consistency. Data were then entered into a secure, password-protected database.

Outcomes

Our primary outcomes were the number of days from index screening abnormality to diagnosis and the proportion of women achieving timely diagnosis (within 30 days or 60 days). The index screening abnormality was determined as the latest date of screening abnormality before the commencement of the diagnostic process.26 The date of diagnosis was determined as the date of definitive tissue diagnosis (biopsy with pathology report) or clinical evaluation indicating no further need for diagnostic evaluation.27 Timely diagnosis was calculated as occurring within 30 days or 60 days from an index abnormality. Because studies have indicated that periods of as few as 90 days from the detection of a breast abnormality to the completion of treatment have clinical significance,28, 29 we chose these periods as optimal indicators of timely diagnosis.

Independent Measures

Most independent variables were used as asked. Age was calculated from birth month and year to the date of the abnormal screening test and was categorized as ages <35 years, 36 to 49 years, 50 to 69 years, and ≥70 years. Site was coded corresponding to the name of the city in which the referring clinic was located. BI-RADS classification of screening mammograms is based on BI-RADS criteria.30 Missed appointments and the presence of comorbidities were collapsed to “0,” reflecting “none,” versus “1 or more.”

Analysis

Analyses were conducted using the SPSS statistical software package (version 19.0; SPSS Inc., Chicago, Ill). We calculated descriptive statistics of group characteristics using chi-square tests, t tests, or log-rank tests, as appropriate. Second, we compared the overall time to diagnosis between navigated participants and control participants using the Kaplan-Meier method and determined mediators associated with time using multivariate Cox regression. Although our follow-up was 365 days (to permit continued navigation beyond diagnosis through treatment, for example), we observed that only 9 participants were diagnosed beyond 180 days. These women were censored beyond that time but were included in proportions of women who had a “timely diagnosis” as undiagnosed in those periods. Other studies of breast cancer diagnosis indicate that this exceeds an adequate follow-up period regarding clinical significance.10, 31 On the basis of the initial results, we stratified our initial time-to-diagnosis measure for navigated and control participants according to BI-RADS category, which was the only independent measure associated with the dependent variable other than navigation. We calculated P values overall and within BI-RADS categories using Cox regression, in which larger hazards ratios (HRs) indicate a faster time to diagnosis. Next, univariate and multivariate odds ratios (ORs)32 were generated to test the association of each characteristic with timely diagnosis within 30 days and 60 days.33 In this analysis, larger ORs indicate greater likelihood of a timely diagnosis. A 2-sided P value < .05 indicated statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Among 480 women who had initial results categorized as BI-RADS-3, BI-RADS-4, or BI-RADS-5, follow-up data were available for 425 women (88.5%) (Table 1). All participants were seen by a primary care clinician in community-based clinics, indicating that, generally, they were uninsured or publicly insured. Rates of missing dates and/or diagnostic status as well as demographic characteristics were invariant between navigated and control participants. In Figure 1, “overall” indicates that the time to diagnosis appears to be lower overall in the navigated group relative to the control group (median, 20 days vs 27 days; HR, 1.32; 95% CI, 1.08-1.62; P = .007). Figure 1 (“BI-RADS-3” and “BI-RADS-4/5”) indicates that the time to diagnosis was significantly shorter only among navigated women with a BI-RADS-3 screening result (median, 9 days vs 40 days; HR, 2.42; 95% CI, 1.63-3.59; P < .001). Subsequent Cox regression analysis revealed that only the number of missed appointments was associated with the time to diagnosis (Table 2).

thumbnail image

Figure 1. Kaplan-Meir time-to-diagnosis comparison, navigated versus control subjects, overall and stratified by BI-RADS. Censored at 180 days.

Download figure to PowerPoint

Table 1. Demographic Characteristics of Women in the Navigated and Control Groups
 No. of Patients (%)a
MeasureControl Group, N = 208Navigated Group, N = 217Total, N = 425
  • Abbreviations: BI-RADS, Breast Imaging Reporting and Data System.

  • a

    Sample sizes vary within cells because of missing or unavailable data.

  • b

    Site names are anonymized in this table.

  • c

    These include any comorbidity noted on a patient's chart or reported by a patient, including: myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, connective tissue disease, peptic ulcer disease, liver disease (mild, moderate/severe), diabetes mellitus (with or without complications), hemiplegia or paraplegia, renal disease, and acquired immunodeficiency syndrome.

Siteb   
 A41 (19.7)33 (15.2)74 (17.4)
 B19 (9.1)18 (8.3)37 (8.7)
 C43 (20.7)50 (23)93 (21.9)
 D38 (18.3)30 (13.8)68 (16)
 E25 (12)41 (18.9)66 (15.5)
 F42 (20.2)45 (20.7)87 (20.5)
Age category, y   
 <3510 (4.8)17 (10.3)27 (7.3)
 36-4975 (36.2)56 (33.9)131 (35.2)
 50-69111 (53.6)91 (55.2)202 (54.3)
 ≥7011 (5.3)1 (0.6)12 (3.2)
Country of origin   
 US12 (17.4)53 (25.5)62 (22.6)
 Mexico14 (20.3)78 (37.5)92 (33.6)
 Other43 (62.3)77 (37)120 (43.8)
Marital status   
 Married/living as married85 (42.1)86 (42.2)171 (42.3)
 Unmarried117 (57.9)116 (57.6)233 (57.7)
Employment status   
 Yes40 (36)89 (42.2)129 (40.1)
 No71 (64)122 (57.8)193 (59.9)
Insurance status   
 Medicare16 (7.8)16 (7.7)32 (7.7)
 Private insurance39 (18.9)37 (17.8)76 (18.4)
 Other local government payment-assistance plan67 (32.5)65 (31.3)132 (31.9)
 None84 (40.8)90 (43.3)174 (42)
BI-RADS category   
 BI-RADS-364 (30.8)73 (33.6)137 (32.2)
 BI-RADS-4114 (54.8)126 (58.1)240 (56.5)
 BI-RADS-530 (14.4)18 (8.3)48 (11.3)
Missed appointments   
 None164 (78.8)179 (82.5)343 (80.7)
 ≥144 (21.2)38 (17.5)82 (19.3)
Comorbidityc   
 No177 (85.1)177 (81.6)354 (83.3)
 Yes23 (11.1)20 (9.2)43 (10.1)
Table 2. Predictors of the Time to Diagnosis for Breast Screening Abnormalitiesa
 95% CI 
MeasureHRLowerUpperP
  • Abbreviations: BI-RADS, Breast Imaging Reporting and Data System; CI, confidence interval; HR, hazard ratio (from Cox proportional regression analysis); Referent, reference category.

  • a

    The time to diagnosis is defined as the time in days from date of the initial abnormal examination to the date of diagnosis. Data were censored at 180 days in this analysis. Nine women were censored because their diagnosis occurred >180 days after initial detection of the abnormality. All covariates were tested in the multivariate model, and only those that were associated significantly with the dependent variable are listed.

Navigated group1.411.151.74.001
BI-RADS    
 Overalld   < .001
 BI-RADS-3 (referent)e
 BI-RADS-4/51.060.842.61.801
Covariatesb    
 Missed appointments: None missed vs any missed (referent)0.470.360.62< .001

In addition, navigated women achieved timely diagnosis significantly more often than women in the control group (30 days: 67.3% vs 57.7%, respectively; P = .045; 60 days: 86.2% vs 78.4%, respectively; P = .023) (Table 3). However, this effect occurred only among women who had an initial BI-RADS-3 screening result (30 days: 83.6% vs 50%, respectively; P < .001; 60 days: 94.5% vs 67.2%, respectively; P < .001). There was no comparable difference among women who had initial BI-RADS-4/5 screening result. Like the time-to-diagnosis analysis, the number of missed appointments was the only significant predictor of a timely diagnosis (Table 4).

Table 3. The Proportion of Participants With a Timely Diagnosis (Within 30 Days and Within 60 Days of Abnormal Screen) Overall and Stratified by Type of Abnormal Screena
 Diagnosed Within 30 DaysDiagnosed Within 60 Days
GroupNo. of Women (%)PNo. of Women (%)P
  • Abbreviations: BI-RADS, Breast Imaging Reporting and Data System.

  • a

    Timely diagnosis is defined as a diagnosis within 30 days and 60 days of the initial abnormal screen, as indicated. Nine women who were censored in the survival models were included here as “not diagnosed” (based on intent to treat).

Overall    
 Navigated, n = 208146 (67.3).045187 (86.2).023
 Controls, n = 217120 (57.7) 163 (78.4) 
BI-RADS-3    
 Navigated, n = 7361 (83.6)< .00169 (94.5)< .001
 Controls, n = 6432 (50) 43 (67.2) 
BI-RADS-4/5    
 Navigated, n = 14485 (59).810118 (81.9).438
 Controls, n = 14488 (61.1) 120 (83.3) 
Table 4. Predictors of Timely Diagnosis Within 30 Days and 60 Days of Breast Abnormalitya
 OverallBIRADS3BIRADS4/5
VariablebOR (95%CI)POR (95%CI)POR (95%CI)P
  • Abbreviations: BI-RADS, Breast Imaging Reporting and Data System; CI, confidence interval; OR, odds ratio (from multiple logistic regression).

  • a

    Timely diagnosis is defined as a diagnosis within 30 days or 60 days of initial abnormal screen. Nine women who were censored in the survival models were included here as “not diagnosed” (based on intent to treat).

  • b

    All potential covariates were tested in multivariate models. Only those associated significantly with the dependent variable are shown.

Diagnosed within 30 d of abnormality      
 Navigation      
  NoReferent Referent Referent 
  Yes1.51 (1.01-2.19).0495.25 (2.36-11.66)< .0010.89 (0.56-1.44).641
 BI-RADS category      
  BI-RADS-31.42 (0.97-2.20).110    
  BI-RADS-4/5Referent     
 Missed appointments      
  NoReferent Referent Referent 
  Yes0.61 (0.42-1.15).7920.57 (0.34-1.21).1940.69 (0.35-1.38).296
Diagnosed within 60 d of abnormality      
 Navigation      
  NoReferent Referent Referent 
  Yes1.72 (1.03-2.85).0388.74 (2.78-27.4)< .0010.90 (0.49-1.67).742
 BI-RADS category      
  BI-RADS-31.41 (0.97-2.22).081    
  BI-RADS-4/5Referent     
 Missed appointments      
  NoReferent Referent Referent 
  Yes0.62 (0.32-1.22).1940.51 (0.32-0.99).0470.90 (0.37-2.20).827

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Time to Diagnosis and Timely Diagnosis Among Navigated Latinas

To our knowledge, this is the first study of Latinas to consider the effect of PN on the time to diagnosis and to stratify those results by the type of initial abnormal screen. In this study, approximately 33% of eligible women had screening results categorized as BI-RADS-3, and the impact of navigating BI-RADS-3 women indicates that they are diagnosed significantly faster than comparable non-navigated women. The finding that this occurs and caused by fewer missed appointments and a more rapid diagnostic confirmation lends support to efforts at reducing the BI-RADS-3 classification to a less threatening BI-RADS result,22, 34 and it has been observed that women literally “suffer” through a BI-RADS-3 result while awaiting diagnosis.32, 35 The current study provides strong implications that navigated Latinas who had a BI-RADS-3 screening result probably were relieved of the burden of worry about their abnormal result.

Patient Navigation: Great Expectations

Studies have demonstrated repeatedly that cancer places an unequal burden on patients with lower socioeconomic status and on racial/ethnic minority women.19, 20, 36, 37 These disparities manifest themselves in lower survival rates among disadvantaged women, and it has been demonstrated that they are the consequence of a sequence of circumstances, including minority status and marginalization, inability to access and adequately use medical resources, unavailability of those resources in some locales, late diagnoses and more severe disease, and similar delays in treatment, ultimately leading to higher rates of death.38, 39 Disparities must be ameliorated, and PN may provide an effective intervention to accomplish this.

PN programs are increasing in the United States despite the lack of empiric evidence for their efficacy or cost-effectiveness.40 The expected impact of PN on some aspects of the cancer care continuum is high, but demonstrating efficacy has been difficult. Evidence was summarized in a recent review, which noted the rapid expansion of PN while underscoring study limitations, including lack of randomization, absence of control groups, small sample sizes, and inability to compare endpoints.41 Although the benefits of applying PN to the barriers faced by low-income, underserved minority groups in dealing with cancer remains unclear, there is some evidence that PN works when applied correctly and in a timely fashion to specific clinical challenges.23 Simultaneously, there is little evidence that we have generally identified those specific challenges and how PN mediates them.40 Therefore, it is difficult to know when, where, and how to apply PN. In the current article, we report 1 successful application of PN to a limited target group (Latinas with a BI-RADS-3 result).

Secular Trends in Screening and Diagnosis and an Unexpected Finding

PN appears to be a specific, rather than general, curative for disparities. Our results indicate an overall decrease in the time to diagnosis between navigated and non-navigated women caused almost exclusively by a decrease in time among women with an initial BI-RADS-3 screen. This may be because secular trends in diagnosing breast cancer among women at some clinics have improved (as in the Mayo Clinic Center for Excellence in Breast Cancer),39 especially when the probable risk of cancer is considered relatively high (eg, BI-RADS-4 or BI-RADS-5). Little further improvement is gained by navigating all women from abnormal screening to diagnosis. Women with an abnormal result of BI-RADS-3, however, may fall into a nosological “no-man's land” of relative less importance—the likelihood of cancer in a woman with a BI-RADS-3 screening result is approximately 2% to 4%.38 This is an unexpected finding: Although we did not design the study to compare the time to diagnosis between BI-RADS groups within the navigated and control arms, the time to diagnosis with the navigation group went from 9 days for those with BI-RADS-3 results to 26 days for those with BI-RADS-4/5 results. This suggests that women who are informed that their screening mammogram is “abnormal” (albeit with a low likelihood of cancer) are 1 group for whom PN, problematically, is appropriate. On the basis of this finding, PN is likely to have a very small impact on breast cancer morbidity and mortality outcomes (ie, if only 2%-4% of BI-RADS-3 women have cancer, then reducing the time to diagnosis will have a very minor population impact on early diagnosis of breast cancer). However, it may have a major impact on addressing Latinas' anxiety associated with abnormal results. Further studies will be required to evaluate these results.

In conclusion, patient-centered navigation to assist Latina women with breast screening abnormalities reduces the time to diagnosis by nearly 66% among Latina women with an initial BI-RADS-3 classification compared with non-navigated controls. Moreover, this reduction is reflected in increases in rates of timely diagnosis within 30 and 60 days if an initial screening abnormality is classified within a range in which the likelihood of cancer is between 2% and 4% (BI-RADS-3). Navigation appears to assist Latinas who have an initial BI-RADS-3 abnormal mammogram in reducing missed follow-up appointments, but not Latinas who have a greater likelihood of a cancer diagnosis.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

This work was supported by Redes En Accion: The National Latino Cancer Research Network (National Cancer Institute grant U01 CA114657-05) and by the Cancer Center Support Group of the Cancer Therapy and Research Center at The University of Texas Health Science Center at San Antonio (P30CA054174).

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES
  • 1
    American Cancer Society. Breast Cancer Facts & Figures 2009-2010. Atlanta, GA: American Cancer Society; 2009.
  • 2
    American Cancer Society. Cancer Prevention & Early Detection Facts & Figures 2009. Atlanta, GA: American Cancer Society; 2009.
  • 3
    Jemal A, Clegg LX, Ward E, et al. Annual report to the nation on the status of cancer, 1975-2001, with a special feature regarding survival. Cancer. 2004; 101: 3-27.
  • 4
    Lantz PM, Mujahid M, Schwartz K, et al. The influence of race, ethnicity, and individual socioeconomic factors on breast cancer stage at diagnosis. Am J Public Health. 2006; 96: 2173-2178.
  • 5
    Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System Survey Data. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2010.
  • 6
    Miller BA, Hankey BF, Thomas TL. Impact of sociodemographic factors, hormone receptor status, and tumor grade on ethnic differences in tumor stage and size for breast cancer in US women. Am J Epidemiol. 2002; 155: 534-545.
  • 7
    Press R, Carrasquillo O, Sciacca RR, Giardina EG. Racial/ethnic disparities in time to follow-up after an abnormal mammogram. J Womens Health (Larchmt). 2008; 17: 923-930.
  • 8
    Elmore JG, Nakano CY, Linden HM, Reisch LM, Ayanian JZ, Larson EB. Racial inequities in the timing of breast cancer detection, diagnosis, and initiation of treatment. Med Care. 2005; 43: 141-148.
  • 9
    Hershman DL, Wang X, McBride R, Jacobson JS, Grann VR, Neugut AI. Delay of adjuvant chemotherapy initiation following breast cancer surgery among elderly women. Breast Cancer Res Treat. 2006; 99: 313-321.
  • 10
    Hershman DL, Wang X, McBride R, Jacobson JS, Grann VR, Neugut AI. Delay in initiating adjuvant radiotherapy following breast conservation surgery and its impact on survival. Int J Radiat Oncol Biol Phys. 2006; 65: 1353-1360.
  • 11
    Gibbons MC. A historical overview of health disparities and the potential of eHealth solutions [serial online]. J Med Internet Res. 2005; 7: e50.
  • 12
    Jemal A, Thun M, Ries L, et al. Annual report to the nation on the status of cancer, 1975-2005, featuring trends in lung cancer, tobacco use, and tobacco control. J Natl Cancer Inst. 2008; 100: 1672.
  • 13
    Viswanath K, Ackerson LK. Race, ethnicity, language, social class, and health communication inequalities: a nationally representative cross-sectional study [serial online]. PLoS One. 2011; 6: e14550.
  • 14
    American Cancer Society. Cancer Facts & Figures 2008. Atlanta, GA: American Cancer Society; 2008.
  • 15
    Howe HL, Carozza S, O'Malley C, Dolecek TA, Finch JL, Kohler B. Cancer in US Hispanics/Latinos, 1995-2000. Springfield, IL: North American Association of Central Cancer Registries; 2003.
  • 16
    United States Bureau of the Census. American FactFinder. Available at: http://factfinder.census.gov/. Accessed August 29, 2010.
  • 17
    Freedman RA, He Y, Winer EP, Keating NL. Trends in racial and age disparities in definitive local therapy of early stage breast cancer. J Clin Oncol. 2009; 27: 713-719.
  • 18
    Ward E, Halpern M, Schrag N, et al. Association of insurance with cancer care utilization and outcomes. CA Cancer J Clin. 2008; 58: 9-31.
  • 19
    Caplan LS, May DS, Richardson LC. Time to diagnosis and treatment of breast cancer: results from the National Breast and Cervical Cancer Early Detection Program, 1991-1995. Am J Public Health. 2000; 90: 130-134.
  • 20
    Chang SW, Kerlikowske K, Napoles-Springer A, Posner SF, Sickles EA, Perez-Stable EJ. Racial differences in timeliness of follow-up after abnormal screening mammography. Cancer. 1996; 78: 1395-1402.
  • 21
    Balleyguier C, Ayadi S, Van Nguyen K, Vanel D, Dromain C, Sigal R. BIRADS classification in mammography. Eur J Radiol. 2007; 61: 192-194.
  • 22
    Giannakopoulou G, Spyrou GM, Antaraki A, et al. Downgrading BIRADS 3 to BIRADS 2 category using a computer-aided microcalcification analysis and risk assessment system for early breast cancer. Comput Biol Med. 2010; 40: 853-859.
  • 23
    Freeman HP, Rodriguez RL. History and principles of patient navigation. Cancer. 2011; 117( S15): 3537-3540.
  • 24
    Allen JD, Shelton RC, Harden E, Goldman RE. Follow-up of abnormal screening mammograms among low-income ethnically diverse women: findings from a qualitative study. Patient Educ Couns. 2008; 72: 283-292.
  • 25
    Mojica CM, Bastani R, Ponce NA, Boscardin WJ. Latinas with abnormal breast findings: patient predictors of timely diagnostic resolution. J Womens Health (Larchmt). 2007; 16: 1468-1477.
  • 26
    Bevers TB, Anderson BO, Bonaccio E, et al. NCCN Clinical Practice Guidelines in Oncology: breast cancer screening and diagnosis. J Natl Compr Canc Netw. 2009; 7: 1060-1096.
  • 27
    Freund KM, Battaglia TA, Calhoun E, et al. National Cancer Institute Patient Navigation Research Program: methods, protocol, and measures. Cancer. 2008; 113: 3391-3399.
  • 28
    Battaglia TA, Roloff K, Posner MA, Freund KM. Improving follow-up to abnormal breast cancer screening in an urban population. A patient navigation intervention. Cancer. 2007; 109( 2 suppl): 359-367.
  • 29
    Palmieri FM, DePeri ER, Mincey BA, et al. Comprehensive diagnostic program for medically underserved women with abnormal breast screening evaluations in an urban population. Mayo Clin Proc. 2009; 84: 317-322.
  • 30
    National Cancer Institute. Breast Cancer Screening (PDQ)—Health Professional. Available at: http://www.cancer.gov/cancertopics/pdq/screening/breast/HealthProfessional. Accessed February 1, 2012.
  • 31
    National Cancer Institute. Cancer Trends Progress Report—2009/2010 Update. Bethesda, MD: National Institutes of Health, Department of Health and Human Services; 2010.
  • 32
    Hosmer D, Lemeshow S. Applied Logistic Regression. New York: Wiley-Interscience; 2000.
  • 33
    Hall FM. Malignancy in BI-RADS category 3 mammographic lesions. Radiology. 2002; 225: 918-920.
  • 34
    Baltzer PA, Benndorf M, Dietzel M, Gajda M, Runnebaum IB, Kaiser WA. False-positive findings at contrast-enhanced breast MRI: a BI-RADS descriptor study. AJR Am J Roentgenol. 2010; 194: 1658-1663.
  • 35
    Keyzer-Dekkera CM, De Vriesb J, van Eschb L, et al. Anxiety after an abnormal screening mammogram is a serious problem. Breast. 2011; 21: 83-88.
  • 36
    Jones BA, Dailey A, Calvocoressi L, et al. Inadequate follow-up of abnormal screening mammograms: findings from the race differences in screening mammography process study (United States). Cancer Causes Control. 2005; 16: 809-821.
  • 37
    Kerlikowske K. Timeliness of follow-up after abnormal screening mammography. Breast Cancer Res Treat. 1996; 40: 53-64.
  • 38
    Eberl MM, Fox CH, Edge SB, Carter CA, Mahoney MC. BI-RADS classification for management of abnormal mammograms. J Am Board Fam Med. 2006; 19: 161-164.
  • 39
    Peres J. Mammography screening: after the storm, calls for more personalized approaches. J Natl Cancer Inst. 2010; 102: 9-11.
  • 40
    Wells KJ, Battaglia TA, Dudley DJ, et al. Patient navigation: state of the art or is it science? Cancer. 2008; 113: 1999-2010.
  • 41
    Paskett ED, Harrop JP, Wells KJ. Patient navigation: an update on the state of the science. CA Cancer J Clin. 2011; 61: 237-249.