Racial Disparities in Immediate Breast Reconstruction After Mastectomy: Impact of State and Federal Health Policy Changes
- First Online:
- Cite this article as:
- Yang, R.L., Newman, A.S., Reinke, C.E. et al. Ann Surg Oncol (2013) 20: 399. doi:10.1245/s10434-012-2607-9
- 248 Views
Federal and Pennsylvania state policies instituted in the late 1990s were designed to improve access to postmastectomy breast reconstruction. We sought to evaluate the impact of these policy changes on access to care among racial minorities.
Mastectomy patients ≥18 years old were identified in the Pennsylvania Health Care Cost Containment Council inpatient database (1994–2004) and classified by immediate breast reconstruction (IBR) status. Rates of IBR were calculated by patient characteristics and year. Patients were stratified by race before (1994–1997) and after (2001–2004) policy changes, and relative odds of IBR were estimated by univariate and multivariate logistic regression analyses with adjustment for known confounders.
Overall rates of IBR were significantly higher in the time period after policy change compared to before policy change (18.5 vs. 32.7 %, p < 0.01). White, black, and Asian patients all saw a significant rise in rates of IBR. However, after adjustment for potential confounders, black patients, Asian patients, and those of mixed or other races all remained less likely to undergo IBR when compared to white patients after policy changes (odds ratio [OR] 0.66, 95 % confidence interval [CI] 0.55–0.80; OR 0.30, 95 % CI 0.18–0.49; OR 0.29, 95 % CI 0.16–0.51, respectively).
Rates of IBR increased across all racial groups after policy changes. However, not all races were affected equally, and thus disparities remained. Future studies are needed to investigate the role of other factors, including cultural preferences in utilization of IBR that might explain residual disparities.
Breast cancer will affect one in eight women during their lifetime, with a 5-year survival rate of 89 %.1 Because of tumor size, personal or family history of breast cancer, or personal preference, many patients choose total mastectomy as treatment for breast cancer. Patients who undergo breast reconstruction after mastectomy experience an improved quality of life when compared with those who do not undergo reconstruction.2,3 Furthermore, immediate breast reconstruction (IBR) has proven successful from a technical and oncologic perspective.4 Despite the published benefits of breast reconstruction, a minority of patients opts to undergo reconstructive breast surgery.5
Early policies aimed at improving access to breast cancer care for underserved women in the United States failed to address an important aspect of breast cancer care: the option for reconstruction after mastectomy.6,7 Health care payers continued to refuse coverage of postmastectomy breast reconstruction on the basis of it being a purely cosmetic procedure.8 Lobbying efforts by dissatisfied clinicians, researchers, policy makers, and breast cancer survivors resulted in the 1997 Pennsylvania Breast Cancer Reconstructive Surgery Coverage Act and the 1998 federal Women’s Health and Cancer Rights Act, which mandated insurance coverage of all breast reconstruction after mastectomy.9 And in 2002, Pennsylvania extended Medicaid coverage of breast reconstruction to all uninsured women diagnosed with breast cancer.9
Despite numerous policy initiatives to guarantee equal access to care for breast cancer patients, there are still reports of disparities in immediate reconstruction amongst racial minorities. More often than white patients, black patients say that breast reconstruction is not discussed or is discouraged.10 Similarly, black patients report that they simply do not know enough about IBR to consent to the operation.10 Black women have been found to have a significantly lower likelihood of reconstruction when compared to white women.11 The impact of policy changes on improving access to care among racial minorities has not been adequately studied.
We aimed to determine the patterns of utilization of IBR during the time period surrounding passage of these legislative acts to examine the influence of policy changes on the likelihood of IBR with particular attention to the issue of racial disparity.
We performed a retrospective cohort study of breast cancer patients 18 years of age or older who underwent an inpatient mastectomy in Southeastern Pennsylvania between 1994 and 2004 and were entered into the Pennsylvania Health Care Cost Containment Council (PHC4) database (Regions 5, 7, 8, and 9). The PHC4 database contains all publically available discharge data from all general acute care hospital discharges in Pennsylvania, including demographics, diagnoses, procedures, and physician and hospital identifiers. The database is maintained by PHC4 to provide health care consultants, researchers, hospital administrators, and health care purchasers insight into best medical practices. The PHC4 database is publically available to researchers for a nominal fee.12
The primary outcome of interest was IBR. Patients were classified as receiving IBR by the presence of ICD-9 procedure codes during the incident admission indicating transverse rectus abdominis myocutaneous free or pedicled flap, latissimus dorsi myocutaneous pedicled flap, deep inferior epigastric perforator free flap, gluteal artery perforator free flap, implant-based reconstruction, or other method of breast reconstruction: 8531, 8532, 8533, 8535, 8550, 8551, 8552, 8553, 8554, 856, 857, 8572, 8573, 8574, 8575, 8576, 8579, 858, 8584, 8585, 8589, 8593, 8594, 8595, or 8596. In the absence of one of these codes patients were classified as not receiving IBR.
All data elements were defined by the Pennsylvania Uniform Claims and Billing Form Reporting Manual.13 Patient characteristics studied included calendar year of mastectomy, race, age, Elixhauser Comorbidity Index, and estimated household income.
Race was obtained directly from the PHC4 database as a categorical variable. Race was initially defined as white, nonwhite, and missing. In subset analysis to evaluate the impact of policy changes on racial disparities, race was further expanded to white, black, Asian and “other.” Hispanics, American Indians, Alaskan Natives, and patients of more than one race were combined into the “other” category as result of small sample sizes. Patients for whom race was missing were excluded from the subset analysis.
The Elixhauser Comorbidity Index is a widely applied tool used to assess comorbidities in administrative discharge databases. The index has been proven both reliable and valid and uses 30 categories of comorbid illness identified using ICD-9 diagnosis codes.14 Patients were classified as having zero, one to two, or greater than two comorbidities using this index. Estimated household income, as determined by median household income of patient ZIP code, was recorded from the PHC4 dataset and stratified by quartile.
Descriptive statistics were performed using the Student’s t test for parametric continuous data and the Fisher’s exact test or the χ2 test, as appropriate, for categorical data. The number of mastectomies and rates of IBR were calculated by year over the study time period. Patients were then classified as undergoing mastectomy before (1994–1997) or after (2001–2004) the policy changes, with the years 1998–2000 excluded from analysis to prevent confounding of data due to the many policy changes that were occurring during those years. Number of mastectomies and rates of IBR were calculated before and after policy changes by patient and hospital characteristics. Percentage increase in rates of IBR was calculated by patient and hospital characteristics. Univariate analysis was used to examine the association between patient and hospital characteristics and IBR. Characteristics found to be significantly associated with IBR in univariate analysis were included in the multivariate model.
A multivariate logistic regression model was developed to estimate the relative odds of IBR versus mastectomy only, before and after the policy changes, with adjustment for calendar year, race, age, and comorbidity index, and estimated household income.
Subset analysis using the expanded classification of races was performed to investigate the impact of policy changes on racial disparities in IBR. Patients in each racial group were classified as undergoing mastectomy before or after the policy changes, and the distribution of individual characteristics was compared by race. Rates of IBR were then calculated by race. The rates of IBR before and after policy changes were compared within each race category using the Student’s t test for independent samples. The difference in reconstruction rates before and after policy changes was calculated within each race category. Univariate logistic regression was used to determine the relative odds of IBR by race before policy changes and after implementation. Multivariate logistic regression was used to assess the likelihood of IBR by racial status with adjustment for age, comorbidity index and estimated household income. Models were constructed with and without an interaction term for race and median household income using the prescribed categories. The likelihood ratio test was used to assess the explanatory power of the models. The results displayed reflect those of the simple model without the interaction term.
Data management was performed by SAS 9.2 (SAS, Cary, NC) and statistical analyses were performed using Stata/SE Version 11.1 (StataCorp LP 2009, College Station, TX). A p value of <0.05 was considered significant for all statistical analyses.
Characteristics and demographics of patients undergoing mastectomy in Pennsylvania, 1994–2004
Before policy changes, 1994–1997
After policy changes, 2001–2004
IBR rate (%)
IBR rate (%)
Year of surgerya
Age at surgerya (years)
Estimated household incomea
After policy changes, there was a 15.1 % rise in rates of IBR among white patients and a 9.4 % rise in rates of IBR among nonwhite patients. The observed increase in the rates of reconstruction differed by income status as well (quartile 1, 10.1 %; quartile 2, 11.2 %; quartile 3, 15.2 %; quartile 4, 15.9 %).
After adjustment for potential confounders, patients were more likely to undergo IBR in the years after the policy changes (odds ratio [OR] 3.98, 95 % confidence interval [CI] 3.41–4.65) when compared to the pre-policy era. Despite the impact of policy changes, nonwhite patients remained less likely to undergo reconstruction when compared to white patients (OR 0.56, 95 % CI 0.48–0.67).
Characteristics of patients undergoing IBR in Pennsylvania by race, before and after state and federal policy changes
Before policy changes, 1994–1997
After policy changes, 2001–2004
White, n (%) (N = 1,660)
Black, n (%) (N = 179)
Asian, n (%) (N = 6)
Other, n (%)b (N = 18)
White, n (%) (N = 2,512)
Black, n (%) (N = 271)
Asian, n (%) (N = 25)
Other, n (%)b (N = 18)
Age at surgerya (years)
Estimated household income ($)a
Subset analysis of trends in the likelihood of undergoing IBR, before and after state and federal policy changes
Before policy changes, 1994–1997
After policy changes, 2001–2004
95 % CI
95 % CI
Age at surgery (years)
Estimated household income
The addition of an interaction term for race and median income did not affect the explanatory power of the models (likelihood ratio test, p > 0.05).
The likelihood of IBR increased significantly over the study time period. However, the percentage rise in the utilization of IBR varied amongst patients with different racial, medical, and economic profiles. Despite policy changes that entitle all breast cancer patients in the United States to insurance coverage of reconstruction after mastectomy, utilization rates remain low especially among racial minorities.
In the period of time after the legislation, utilization of IBR increased for all of the racial groups studied. However, disparities were not entirely eliminated. All racial minorities still had lower rates of IBR when compared to white patients after policy changes. Furthermore, after adjustment for multiple confounders, we found that black patients, Asian patients and patients from the “other” race category remained less likely to undergo IBR. These findings prompt further thought as to why the state and federal health policy changes have not yet eliminated racial disparities in access to breast cancer care.
Low socioeconomic status, lack of health insurance, and low literacy among racial minorities are thought to contribute to disparities in treatment of cancer.15–19 We similarly found that patients with higher estimated household income were significantly more likely to undergo IBR when compared to those with lower estimated household income. The availability of Medicaid coverage of postmastectomy breast reconstruction for all uninsured Pennsylvanian women should have eliminated disparities in IBR amongst lower income patients without medical contraindications. Unfortunately, the dissemination of information regarding these policies may have variable penetrance related to socioeconomic status.
Another potential barrier to policy effects is language. Of Asians living in the United States, the percentage of those who do not speak English at home is 55 % of Vietnamese, 46 % of Chinese, and 22 % of Filipinos.20 Similar to what was reported in a study of Surveillance, Epidemiology, and End Results data from 1998, we found that Asian woman were approximately one-third as likely to be reconstructed as white patients after adjustment for potential confounders.11 It is also possible that cultural value of women’s breasts may differ by race. Nevertheless, this appears to be an opportunity for the healthcare system to identify strategies to educate all women about breast cancer reconstruction, despite race, ethnicity or primary language spoken.
We report the annual rate of IBR to be between 19 and 33 % per year. The American Society of Plastic Surgeons estimated that 39–42 % of all breast reconstructions performed by their society members are immediate, while the remainder are delayed.9 Therefore, it is possible that the true rate of breast reconstruction, or immediate combined with delayed reconstruction, is more than double the rate that we found in our study. However, another study showed that despite extensive counseling regarding postmastectomy breast reconstruction and the opportunity for preoperative consultation with a plastic surgeon, still only 21 % of mastectomy patients elected for reconstruction.21 Further studies are needed to determine what factors influence individual patient preference and how this affects a patient’s likelihood to undergo reconstruction.
We report increasing rates of IBR over the study time period. We note that this increase was not merely the result of a decrease in absolute number of mastectomies performed, with a stable number of breast reconstructions. Rather, there was an increase in the number of IBRs performed and a slight decrease in number of inpatient mastectomies performed. This is in keeping with another single center study evaluating mastectomies that reported a decrease in the rate of mastectomies performed during the years 1997–2003.22 Other studies have reported an increase in total number of mastectomies performed during recent years.23 The increase in mastectomy rates were attributed to improved breast imaging techniques, more widespread use of genetic testing, greater availability of reconstruction techniques, and trends in patient preference.24–28 One possible explanation for our contrary finding is that an increasing number of mastectomies are being performed in the outpatient setting. It is possible that our study of inpatient admissions underestimated the total number of mastectomies performed and that reconstruction rates might be lower when considering all mastectomy patients. One recent study of Medicare patients reported a significantly lower rate of breast reconstruction after outpatient mastectomy when compared to inpatient mastectomy, with racial disparities more pronounced among inpatient mastectomy patients.29
Our retrospective study is limited in the ability to show a direct effect of policy changes on IBR. We can only demonstrate an association between timing of the policy changes and the increase in the use of IBR, with extensive adjustment for potential confounders. We must not overlook the possible effect of other changes that occurred during the study time interval that could have impacted likelihood of patients to undergo breast reconstruction. These changes include improved physician and patient awareness regarding the psychological benefits of breast reconstruction, more widespread use of autologous reconstruction and thus additional reconstructive options available, and enhanced screening methods diagnosing more early stage breast cancers eligible for reconstruction.27,30,31 Unfortunately we were not able to control for these confounding variables with our study design. Furthermore, although we excluded the years 1998–2000 from our analysis as a result of the many policy changes that occurred during those years, there may have been confounding effects resulting from the continued policy changes, such as extension of Medicaid coverage, implemented after the year 2000.
In our study, we could not determine how each patient was diagnosed with breast cancer. It is possible that patients diagnosed through coverage from federal and state-funded screening programs would have higher rates of reconstruction regardless of their economic means, simply as a result of their overall awareness of public support for breast cancer patients. Furthermore, while we were able to adjust for the burden of patient comorbidities, we could not examine the effect of tobacco usage, body habitus, previous radiation, and prior surgery that might have limited the possibilities for IBR involving autologous tissue transfer.32,33 However, because we looked at both implant-based and autologous reconstruction, it is unlikely that these effects substantially affected our results.
In conclusion, we found a significant rise in rates of IBR throughout the study time period. Even after adjustment for multiple confounders including mastectomy year, patients remained almost four times more likely to undergo IBR after the passage of policy changes then they were before the policy became law. The legislation seemed to improve access to care among all individuals in Southeastern Pennsylvania; however, remaining racial disparities in IBR provide evidence that there still is room for improvement. Future studies are needed to understand the individual contributions of patient and provider preferences to these persistent disparities in IBR.
The PHC4 is an independent state agency responsible for addressing the problem of escalating health costs, ensuring the quality of health care, and increasing access to health care for all citizens regardless of ability to pay. PHC4 has provided data to this entity in an effort to further PHC4’s mission of educating the public and containing health care costs in Pennsylvania. PHC4, its agents, and staff, have made no representation, guarantee, or warranty, express or implied, that the data—financial, patient, payor, and physician specific information—provided to this entity, are error-free, or that the use of the data will avoid differences of opinion or interpretation. This analysis was not prepared by PHC4. This analysis was done by the authors. PHC4, its agents and staff, bear no responsibility or liability for the results of the analysis, which are solely the opinion of this entity.
Conflicts of interest
The authors declare no funding or conflicts of interest.