Currently, breast imaging practices use variable postsurgical surveillance protocols without uniform recommendations. This can lead to lapses in patient adherence, which has the potential to impact clinical outcomes. The purpose of this study is to evaluate the relationship between sociodemographic factors and postsurgical surveillance imaging protocols on patient adherence rates. Results show that sociodemographic disparities exist as barriers for annual mammography surveillance in patients after breast-conserving surgery.
Objective:
Currently, national and international breast imaging practices utilize variable postsurgical surveillance protocols without uniform recommendations. Because of the innate differences between screening versus diagnostic mammography from scheduling flexibility to out-of-pocket costs, this creates the opportunity for lapses in patient adherence, which has the potential to impact clinical outcomes. The purpose of this study is to evaluate the relationship between sociodemographic factors and postsurgical surveillance imaging protocols on patient adherence rates.
Methods:
This retrospective study reviewed 3 years of surveillance imaging for all patients having breast-conserving surgery at our institution from January 2011 to December 2016. Follow-up adherence was defined as returning for all of the first 3 years of annual follow-up after breast-conserving surgery (institutional surveillance protocol). Associations between adherence to surveillance imaging and patient sociodemographic characteristics were evaluated using univariate and multivariate logistic regression.
Results:
The study included 1082 patients after breast surgery, 715 of whom adhered completely to the first 3 years of annual follow-up (66.1%). Black women were 1.36 times less likely to follow up annually compared with White women (95% confidence interval 1.02-1.80). Similarly, patients with Medicare were 1.84 times less likely to follow up annually compared with patients with private insurance (95% confidence interval 1.34-2.51). Women with benign breast disease after breast-conserving surgery were significantly less likely to adhere to annual surveillance than women with breast cancer.
Conclusion:
Sociodemographic disparities exist as barriers for annual mammography surveillance in patients after breast-conserving surgery.
As of February 2021, more than 3.8 million women in the United States are breast cancer survivors.1 In the past few decades, mortality rates for patients with breast cancer have decreased because of advancements in breast cancer treatment and increased early detection through screening mammography.2 However, because the lifetime risk for developing local recurrence can be as high as 15% in these women, it is important to emphasize the importance of consistent imaging surveillance follow-up.3 Surveillance imaging can detect both local recurrences after breast-conserving surgery (BCS) and new cancer diagnoses in the either breast.4
Previous literature demonstrates that annual surveillance mammography has the potential to reduce patient mortality from breast cancer recurrence by 20% to 47%.5- 7 For patients with benign breast disease after BCS, current guidelines recommend returning to routine annual screening mammography.8 Currently, national breast imaging practices utilize variable post-BCS surveillance protocols without uniform recommendations for patients whose pathology resulted in breast cancer.9 This ranges from immediate return to annual screening mammography to indefinite use of diagnostic mammography, with 1 to 2 years of diagnostic mammography being the most common.9 Innate differences between screening versus diagnostic mammography from scheduling flexibility to out-of-pocket costs create the opportunity for lapses in patient adherence, which in turn has the potential to impact clinical outcomes.3,10
To maximize the benefits of postsurgical surveillance imaging protocols in women with a personal history of breast cancer or benign breast disease, it is vital that these women consistently return for annual mammography. Thus, investigating if sociodemographic factors influence patient follow-up adherence in the setting of postsurgical surveillance is relevant and may expose underlying disparities, which could be targeted to improve the quality of patient care. The aim of this study is to evaluate the relationship between sociodemographic factors and patient adherence to 3 years of mammography follow-up after BCS for breast cancer or benign breast disease.
Study Setting
At our institution, the institutional review board reviewed this study and designated it as exempt under the Department of Health and Human Services regulations. This study is HIPAA compliant.
Our institution is an academic medical center that provides tertiary and quaternary patient care. Additionally, many of our patients come from different states and countries to receive their initial surgical care at our institution and continue their follow-up imaging care closer to their origin residence. Our institution's current surveillance imaging protocol for patients with a history of BCS for breast cancer treatment is annual diagnostic mammography follow-up for the first 3 years, returning to annual screening mammography thereafter. Our protocol follows the recommendations by the ACR Appropriateness Criteria for surveillance of patients with asymptomatic breast cancer after BCS.11,12 After BCS for a benign breast disease, women at our institution are recommended to return directly to annual screening mammography.8
Subjects and Study Design
A retrospective cohort study was performed to review the data from all BCSs performed at our institution between January 1, 2011, and December 31, 2016. This period was chosen to ensure that our data included 3 full years of annual follow-up for each patient in the cohort. This patient data set overlaps with our previous study investigating the frequency of additional views utilized in annual diagnostic mammography surveillance follow-up.13
Current Procedural Terminology codes were used to identify mammographic and ultrasound-guided localizations for lumpectomies performed during our study period. Our institution's value analytics team then extracted the corresponding patient medical record numbers from the localization imaging from our institution's electronic medical record, Epic (Epic Systems Corporation, Verona, Wisconsin). The final pathologic diagnosis of the surgical specimen was subsequently extracted. Manual chart review was performed for the medical record numbers without an electronically extracted final pathologic diagnosis to complete the final data set. Patients were categorized as having cancer if the final pathologic diagnosis yielded ductal carcinoma in situ or an invasive breast carcinoma. Any surgical specimen that yielded a noncancerous breast diagnosis was recorded as benign.14 Potential covariates were also extracted from Epic including age at the time of the examination, race, employment status, zip code, insurance status, and molecular subtype. The first insurance type listed for each patient was used in the analysis. By using the University of Wisconsin Neighborhood Atlas, patient zip codes that were extracted from our electronic medical record were used to determine the National Area Deprivation Index as a measure of disparity.15 The demarcation threshold used in our study was the top 15% of disadvantaged neighborhoods because this patient population is associated with higher rates of rehospitalizations.16
Patients who returned for at least one of their annual follow-up examinations (screening or diagnostic) were included in our study. We defined complete patient follow-up adherence as returning for all 3 subsequent years of follow-up after surgery. We defined year 1 as a follow-up examination performed within 13 months from the date of surgery. We defined year 2 as a follow-up examination performed either between 13 and 24 months from the date of surgery or between 11 and 13 months from the prior examination. We defined year 3 as a follow-up examination performed either between 25 and 36 months from the date of surgery or between 11 and 13 months from the prior examination. Diagnostic mammography examinations for symptomatic patients were excluded during manual chart review. Patients who did not have a single screening or diagnostic imaging follow-up within 3 years of surgery were excluded.
Statistical Analysis
All statistical analyses were performed using computing software program R 2017 (R Foundation for Statistical Computing, Vienna, Austria). The primary outcome of interest was adherence to the surveillance imaging protocol (annual screening or diagnostic mammogram for 3 years). Associations between adherence to surveillance imaging and patient sociodemographic and imaging characteristics were evaluated using univariate and multivariate logistic regression, with odds ratios and 95% confidence intervals presented.
Study Characteristics
A total of 1959 patients were identified as having undergone a BCS during the study period. Figure 1 depicts the inclusion and exclusion of patients from this study. Patients who never returned for follow-up at our institution, those who had a mastectomy during the follow-up period, those who died during the follow-up period, and those under the age of 40 were excluded. A total of 1082 unique patients were ultimately included in this study. Table 1 presents the baseline characteristics of the patients included in this study. Of the 1082 patients included, 715 (66.1%) patients were adherent to the first 3 years of mammographic surveillance after their surgery.
Table 1. Baseline characteristics of patients included in the study
| Characteristic | n = 1,082* |
|---|
| Age | 59 (51–68) |
| Race | |
| White or Caucasian | 674 (64.7%) |
| Asian or Pacific Islander | 71 (6.8%) |
| Black or African American | 296 (28.4%) |
| Unknown | 41 |
| Employment status | |
| Full time | 391 (37.2%) |
| Not working | 164 (15.6%) |
| Other | 96 (9.1%) |
| Retired | 399 (38.0%) |
| Unknown | 32 |
| Insurance type | |
| Private | 672 (70.4%) |
| Medicare | 238 (24.9%) |
| Medicaid | 23 (2.4%) |
| Self-pay | 21 (2.2%) |
| Unknown | 128 |
| Area Deprivation Index | |
| <85% | 963 (91.0%) |
| ≥85% | 95 (9.0%) |
| Unknown | 24 |
| Surgery performed for breast cancer | 678 (62.6%) |
| 3-y follow-up | |
| Complete follow-up | 715 (66.1%) |
| Incomplete follow-up | 367 (33.9%) |
Effect of Patient Characteristics on Follow-up Adherence to Imaging Surveillance Protocol
Table 2 shows the association between patient characteristics and adherence to annual surveillance. Age, race, insurance type, employment type, and reason for surgery were significantly associated with adherence to surveillance protocol on univariate analysis. On multivariate analysis, employment status, insurance type, and reason for surgery remained statistically significant.
Table 2. Univariate and multivariate logistic regression analysis for adherence to annual surveillance after breast-conserving surgery
| Characteristic | Summary Statistics* | Univariate Analysis | Multivariate Analysis |
|---|
| Complete Follow-up, n = 715 | Incomplete Follow-up, n = 367 | OR | 95% CI | P Value | OR | 95% CI | P Value |
|---|
| Age | 61 (52–68) | 56 (48–67) | 0.98 | 0.97–0.99 | .002 | 1.01 | 0.99–1.03 | .4 |
| Race | | | | | | | | |
| White or Caucasian | 461 (66.3%) | 213 (61.6%) | — | — | | — | — | |
| Asian or Pacific Islander | 52 (7.5%) | 19 (5.5%) | 0.79 | 0.45–1.35 | .4 | 0.55 | 0.25–1.15 | .13 |
| Black or African American | 182 (26.2%) | 114 (32.9%) | 1.36 | 1.02–1.80 | .036 | 1.34 | 0.91–1.95 | .14 |
| Unknown | 20 | 21 | | | | | | |
| Employment status | | | | | | | | |
| Full time | 248 (35.6%) | 143 (40.4%) | — | — | | — | — | |
| Not working | 100 (14.4%) | 64 (18.1%) | 1.11 | 0.76–1.61 | .6 | 0.92 | 0.55–1.52 | .7 |
| Other | 62 (8.9%) | 34 (9.6%) | 0.95 | 0.59–1.51 | .8 | 0.56 | 0.27–1.10 | .10 |
| Retired | 286 (41.1%) | 113 (31.9%) | 0.69 | 0.51–0.92 | .013 | 0.48 | 0.29–0.82 | .007 |
| Unknown | 19 | 13 | | | | | | |
| Insurance type | | | | | | | | |
| Private | 498 (74.3%) | 174 (61.3%) | — | — | | — | — | |
| Medicare | 145 (21.6%) | 93 (32.7%) | 1.84 | 1.34–2.51 | <.001 | 2.79 | 1.73–4.56 | <.001 |
| Medicaid | 14 (2.1%) | 9 (3.2%) | 1.84 | 0.75–4.27 | .2 | 1.88 | 0.64–5.17 | .2 |
| Self-pay | 13 (1.9%) | 8 (2.8%) | 1.76 | 0.69–4.25 | .2 | 3.11 | 0.63–12.5 | .12 |
| Unknown | 45 | 83 | | | | | | |
| Area Deprivation Index | | | | | | | | |
| <85% | 639 (91.0%) | 324 (91.0%) | — | — | | — | — | |
| ≥85% | 63 (9.0%) | 32 (9.0%) | 1.00 | 0.63–1.55 | >.9 | 0.92 | 0.51–1.62 | .8 |
| Unknown | 13 | 11 | | | | | | |
| Reason for surgery | | | | | | | | |
| Benign breast disease | 163 (22.8%) | 241 (65.7%) | — | — | | — | — | |
| Breast cancer | 552 (77.2%) | 126 (34.3%) | 0.15 | 0.12–0.20 | <.001 | 0.14 | 0.10–0.20 | <.001 |
This study demonstrated that racial and financial disparities exist as barriers for annual mammographic surveillance in patients after BCS. Specifically, Black women and women without commercial insurance were significantly less likely to consistently return for annual surveillance. In addition, having surgery for benign breast disease was associated with decreased adherence compared with having surgery for breast cancer.
Being of Black race was found to be a predictor of nonadherence to our 3-year diagnostic surveillance protocol compared with being of White race. This is an important observation because Black women have a significantly higher risk of breast cancer recurrence and overall higher mortality rate compared with White women.17- 19 Furthermore, prior studies have shown that Black women are less likely to obtain adequate follow-up for their breast care compared with White women.20 This implies that earlier and consistent imaging evaluation for these women may impact survival outcomes.
Independent of race, women with Medicare were significantly less likely to consistently return for annual surveillance compared with women with commercial insurance. Currently, patients without insurance coverage can often obtain a screening mammogram without charge through many organizations such as the Susan G. Komen Breast Cancer Foundation.21- 23 Furthermore, women undergoing screening mammography covered under the Affordable Care Act and Medicare have no deductible and no out-of-pocket copay.22 However, with diagnostic mammography, patients are more likely to incur out-of-pocket costs, which can vary greatly depending on insurance type and state.24 The average cost of a diagnostic mammogram with commercial insurance is $349 ± $492 and can be up to $486 ± $627 if a breast or axillary ultrasound is also obtained during the examination.24 Surprisingly, this is only marginally decreased for patients with Medicare and Medicaid: $297 ± $518 and $373 ± $627, respectively.24 Previous study by The Martec Group for Susan G. Komen demonstrated that regardless of insurance status, patients are frustrated and confused with the costs of diagnostic imaging leading to the cancellation, postponement, or only returning for follow-up if the patient felt it is was necessary.25 Barriers to health care access exist specifically for patients with lower incomes and for racial minorities.26 Furthermore, the 5-year relative survival rates for patients with breast cancer are lower in low-income patients compared with higher-income patients.27 This emphasizes the associated financial burden that annual diagnostic surveillance can represent for these patients and the potential impact that this can have on patient outcomes. Therefore, despite understanding the importance of annual surveillance with mammography, because of the high cost if performed as a diagnostic examination, patients may actively decide not to follow up after BCS, which could put their survival at risk. Additionally, the option to return to annual screening mammography after the first year of diagnostic mammography follow-up in the setting of post-BCS surveillance without a difference in missed recurrent cancers may allow for increased patient adherence by reducing these financial barriers.13 Although the average cost for patients with commercial insurance is slightly higher than Medicare patients, these patients often have higher incomes compared with patients with Medicare (who are over age 65 and more likely retired). This suggests that patients with Medicare may have difficulty paying for the diagnostic examination contributing to the lower adherence to follow-up.
We also examined whether the reason for surgery (breast cancer versus benign breast disease) impacted patient follow-up adherence. In our study, women who had surgery that resulted in benign breast disease were significantly less likely to be adherent to the recommended annual mammography follow-up compared with women with a cancer diagnosis. This is an important observation because benign breast disease is an established risk factor for breast cancer.28,29 Particularly, women with diagnoses of lobular neoplasms (lobular carcinoma in situ and atypical lobular hyperplasia) and atypical ductal hyperplasia have an increased lifetime risk of developing breast cancer in the future.30- 32 The ACR considers these women intermediate risk (lifetime risk of breast cancer 15%-20%), which is similar to women with a personal history of breast cancer.33 Further exploration of the underlying causes resulting in this decrease in adherence for this specific patient population is required moving forward.
One substantial benefit of diagnostic mammography is that patients receive their results at the end of the appointment instead of waiting to receive their results in the mail or potentially being recalled for a follow-up appointment as could happen with screening mammography.34 The wait time between notification of a screening recall and the diagnostic follow-up appointment has been shown to be a source of short-term patient anxiety and presents a risk of becoming lost to follow-up.21,35,36 The benefits of screening mammography, however, are increased availability and flexibility of appointments for patients and often reduced out-of-pocket costs compared with diagnostic mammography.10 Therefore, these drawbacks of diagnostic mammography may also represent barriers to health care access for some patients resulting in patient nonadherence as evident by our results. Furthermore, the option of same-day interpretation of screening mammography exists and has been met with increased patient satisfaction.37,38 Our findings suggest that the benefits of screening mammography may outweigh the primary benefit of diagnostic mammography (immediate results) and should be considered, especially for practices currently utilizing prolonged or indefinite durations of diagnostic mammography surveillance follow-up.9
Limitations in our study are as follows. This study was performed at a single multisite academic institution, and results may not be generalizable to other settings. Patients who did not present consistently for annual follow-up at our institution could have had their subsequent imaging surveillance at a different institution, which may have affected our results. Furthermore, other potential covariates were not available in our electronic medical record, which could affect our results, including patient relocation during the surveillance period, declining diagnostic mammography services, whether patients had a primary care provider, and treatment intensity. The size of the study is relatively small for subset analyses, and larger studies are warranted to confirm our findings.
In conclusion, the advantages and disadvantages of diagnostic mammography must be considered when making imaging surveillance recommendations for breast cancer survivors after BCS. This is particularly important given our findings that racial and financial disparities may represent barriers for these patients to obtain consistent annual follow-up if multiple consecutive years of diagnostic mammography are requested. Furthermore, women with benign breast disease after BCS were significantly less likely to adhere to annual surveillance than women with breast cancer, despite the fact that these women still have an intermediate lifetime risk of developing breast cancer in the future. We are optimistic that this study may provide evidence to support the return to annual screening mammography sooner in practices' postsurgical surveillance protocols for breast cancer survivors and encourage more targeted retention interventions for women with benign breast diseases. This could not only result in improving the quality of individualized follow-up surveillance after BCS but also present an opportunity to close disparity gaps preventing patient follow-up with annual mammography recommendations.
Women with benign breast disease after BCS were significantly less likely to adhere to annual surveillance than women with breast cancer.
Black women were significantly less likely to return annually for diagnostic mammography surveillance imaging in the setting of BCS for breast cancer treatment.
Women with Medicare were significantly less likely to return annually for diagnostic mammography surveillance imaging in the setting of BCS for breast cancer treatment.
Targeted retention interventions may be required to improve adherence to annual mammography recommendations for women with benign breast diseases after BCS.
Returning to annual screening mammography sooner offers the potential to reduce the financial and availability barriers that follow-up diagnostic mammography surveillance presents for patients after BCS.