Correlates of breast reconstruction

Results from a population-based study


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



Immediate or early postmastectomy breast reconstruction is performed infrequently. To the authors' knowledge, little is known regarding surgeon or patient perspectives on reconstruction treatment decisions. The purpose of the current study was to identify patient attitudes and preferences associated with breast reconstruction, and whether these differed by race.


A sample of women age ≤ 79 years who were diagnosed with ductal carcinoma in situ and invasive breast carcinoma between December 2001 and January 2003 was identified from the Surveillance, Epidemiology, and End Results (SEER) registries of Detroit and Los Angeles. Eligible subjects completed a questionnaire at a mean of 7 months after diagnosis. The Wald chi-square test and logistic regression were used for data analysis.


Of the 1844 respondents, 646 underwent a mastectomy (35.0% of the total sample) and 245 of these patients received breast reconstruction (38.0%; of the mastectomy group). On multivariate analysis, younger patient age, higher educational levels, and earlier stage of disease were found to be significantly associated with breast reconstruction. Although 78.2% of women reported that breast reconstruction was discussed, only 11.2% correctly answered 3 basic knowledge questions regarding the procedure. The desire to avoid more surgery was the most common reason for not undergoing breast reconstruction.


The results of the current study found that the majority of women were aware of breast reconstruction but choose not to undergo the procedure. Lack of knowledge and a greater perception of barriers to the procedure were more common among African-American patients and women with a lower education level, suggesting a need for improved educational strategies. Cancer 2005. © 2005 American Cancer Society.

The switch from radical mastectomy to modified radical mastectomy coupled with advances in plastic surgical techniques has made immediate breast reconstruction an option for the majority of patients who undergo mastectomy. Despite this, relatively few women receiving mastectomy undergo immediate or early (within 3 months of their diagnosis of breast carcinoma) breast reconstruction. In a study using a large convenience sample from the National Cancer Database, Morrow et al. found that only 8.3% of 68,348 women undergoing mastectomy between 1994–1995 had immediate or early breast reconstruction.1 A population-based study using 1998 data from the Surveillance, Epidemiology, and End Results (SEER) Registry Program found that 15% of mastectomy patients underwent breast reconstruction within 4 months of mastectomy.2 Both of the studies demonstrated variations in the use of reconstruction on the basis of age, race, and geographic region.1, 2

It has been suggested that large variations in patterns of surgical treatment for breast carcinoma are evidence of the failure to involve women in the surgical decision-making process.3, 4 However, to our knowledge, little is known regarding the way decisions regarding breast reconstruction are made. To our knowledge, mastectomy with breast reconstruction has never been compared with mastectomy alone in a prospective, randomized trial, and data regarding cancer outcomes are derived primarily from single-institution studies, the majority of which are retrospective.5–7 Low rates of early breast reconstruction could reflect surgeons' concerns that immediate or early reconstruction might delay the use of postoperative systemic therapy or hinder the detection of local disease recurrence. Alternatively, low rates of breast reconstruction could be an indication of either patient reluctance to undergo an extra surgical procedure at a time when they are coping with a new diagnosis of a potentially life-threatening disease, or a lack of awareness of the availability of reconstruction.

To improve our understanding of breast reconstruction treatment decisions, we conducted a large, population-based survey of women who were recently diagnosed with breast carcinoma in Detroit and Los Angeles. These cities were chosen because they are multiethnic urban environments that have population-based cancer registries able to identify most women with breast carcinoma shortly after diagnosis. The objectives of the current study were to explore patient perceptions about involvement in the breast reconstruction decision, to determine whether particular patient attitudes and preferences were associated with the receipt of reconstruction, and to determine whether involvement in the decision to undergo breast reconstruction and attitudes toward the procedure differed by race.


Study Population

Women age ≤ 79 years who were diagnosed with ductal carcinoma in situ (DCIS) and invasive disease and identified by the SEER Registries of the greater metropolitan areas of Detroit and Los Angeles during a 14-month period between December 2001 and January 2003 were eligible for the current study.

Database and Sampling

Investigators from both registries identified the study sample and implemented the mailing of a self-administered survey based on a uniform protocol. Women with breast carcinoma were identified and initial pathology reports were collected within 6 weeks of diagnosis for 90% of patients in Detroit and nearly 100% of patients in Los Angeles. Eligible patients underwent a definitive surgical procedure, resided in the catchment area of the SEER site, and were able to complete a questionnaire in English or Spanish. All Asian women and all U.S.-born women age < 50 years who were diagnosed with invasive disease in Los Angeles during our study period were excluded because these women already were being enrolled in other studies. Women with a diagnosis of lobular carcinoma in situ were excluded because the natural history of and recommended treatment for this diagnosis differ from that for DCIS.

We prospectively selected all patients of DCIS and a random sample of invasive patients meeting the study criteria (oversampling African-American women) each month into the preliminary study sample (n=2647). Approximately 90% of all accrued patients were eligible for the study (n = 2382). The survey was completed by 77.4% of eligible patients (92.4% of whom completed a written survey and 7.6% of whom completed an abbreviated telephone survey [n = 1844]). Compared with survey respondents, nonrespondents were of a similar age, but were less likely to be white (69.4% vs. 76.6%; P < 0.001), were more likely to have American Joint Committee on Cancer (AJCC) Stage II disease (25.2% vs. 20.4%; P = 0.034), and were more likely to have undergone a mastectomy (34.7% vs. 30.0 %; P = 0.021). The sample selected for the analyses in the current study was the 646 respondents (35.0%) who underwent a mastectomy.

Data Collection and Management

Physicians were notified of our intent to contact patients. An introductory letter was sent to all potential subjects approximately 3 months after diagnosis followed by a telephone call to assess eligibility. A questionnaire and gift worth $10 were mailed to all eligible women who agreed to participate and to people who could not be reached by telephone (approximately 14% of potential respondents) on a monthly basis. The Dillman survey method was used to encourage response.8

SEER clinical data from hospital-based sources were merged with survey data for 98.2% of patients. The study protocol was approved by the Institutional Review Boards of the University of Michigan, the University of Southern California, and Wayne State University.


The main dependent variable was the receipt of reconstruction, constructed as a dichotomous variable. Self-report of local therapy, including receipt of breast reconstruction after mastectomy, was used in all patients except those in which the self-report was ambiguous or missing (n=26), in which case SEER data were used. Similar to other studies,9 we found that self-report of definitive surgical treatment and SEER data yielded the same surgical procedure for 96.3% of patients in the current study sample. Excluding those patients in which there was a conflict between the self-report and the SEER data did not change the results. Patients who had started or finished breast reconstruction were classified as having undergone reconstruction. Those who were planning, considering, or not considering breast reconstruction were coded as not having undergone breast reconstruction.

The principle independent variables were age at diagnosis (age < 50 yrs, 50–64 yrs, and ≥ 65 yrs), race (white, African American, and other ethnicity), education (some high school, high school graduate, some college, college graduate, and unknown), income (< $20,000, $20,000– $69,999, ≥ $70,000, and unknown), marital/domestic partner status (currently living with partner or not), disease stage, number of comorbidities (0, 1, or 2 or more as derived from a list in the questionnaire of 6 common medical comorbidities), and the presence of a clinical contraindication to breast-conserving treatment. The summary cancer stage was classified using the AJCC TNM staging system for breast carcinoma (DCIS or invasive carcinoma of Stages I–III).10

Patients were asked whether their surgeon explained breast reconstruction to them or referred them to a plastic surgeon for a discussion about reconstruction. Patients who reported that they were not considering breast reconstruction were asked to identify the reasons for the decision from a checklist that included choices such as not wanting to undergo more surgery, not feeling that a new breast was important, a lack of availability of breast reconstruction locally, or concerns about interference with the detection of recurrent breast carcinoma at the surgical site. Finally, patients were asked three true-false questions to determine their level of knowledge about breast reconstruction. The items were: 1) breast reconstruction often involves more than one surgery (true); 2) having breast reconstruction can make it difficult to determine whether breast disease has returned to a breast (false),11, 12 and 3) a breast that has been reconstructed has sensation or feeling (false).

Statistical Analysis

We calculated the proportions of patients who underwent breast reconstruction using sociodemographic and clinical variables. The Wald chi-square test was used to test for bivariate associations between breast reconstruction and each of the independent variables. Sample weights were included to adjust for design effects resulting from our differential sampling across ethnicity and stage. We then used logistic regression to calculate adjusted odds ratios for the association between the receipt of breast reconstruction and independent variables. Second-order interactions between selected covariates were evaluated, particularly geographic site and other covariates, but none were found. We also examined patient reports of their communication with their surgeon(s) about breast reconstruction, the patients' knowledge about breast reconstruction, and their reasons for not undergoing reconstruction. Point estimates were adjusted for design effects using a sample weight that accounted for differential selection by stage, ethnicity, and nonresponse. All analyses were performed using SAS software (version 8.2; SAS Institute Inc., Cary, NC).


Sample Characteristics

Table 1 shows characteristics of the sample patient distribution. The mean patient age was 58.3 years, and 64.3% of the patients were white. Approximately 54.8% had some college or were college graduates and 20.6% had family incomes of ≥ $70,000. Of the patients studied, 42.1% had DCIS or Stage I invasive breast carcinoma, and 53.4% had 1 or more comorbidities. Table 1 also shows the distribution of the receipt of breast reconstruction by variable categories. Compared with patients who did not undergo breast reconstruction, patients who did undergo reconstruction were more likely to be younger and white, and to have some college education, a higher income, and an earlier stage of disease.

Table 1. Sample Characteristics
Variablen = 646Patient distribution (%)Received reconstruction (%)P value
  • SD: standard deviation; AJCC: American Joint Committee on Cancer. Percentages are weighted to account for differential selection by stage of disease, ethnicity, and nonresponse.

  • P values test differences in the receipt of breast reconstruction for each set of variables.

  • a

    Unknowns (n = 6) not included.

  • b

    Unknowns (n = 2) not included.

Study site    
 Los Angeles29444.728.10.002
Age (mean, 58.3 yrs; SD, 11.8)    
 < 50 yrs16524.153.0<0.001
 50–64 yrs28143.836.6 
 ≥ 65 yrs20032.18.3 
 Some high school9315.215.2<0.001
 High school graduate12420.220.0 
 Some college21332.235.5 
 College graduate15122.648.8 
 < $20,00014022.614.3<0.001
 ≥ $70,00013920.656.8 
Married/domestic partnerb    
AJCC stage    
No. of comorbidities    
 ≥ 218127.716.7 

Factors Associated with Breast Reconstruction

Table 2 shows the independent associations between covariates and receipt of breast reconstruction based on a logistic regression model. Younger age, white race, and earlier stage of disease remained significant correlates of the receipt of breast reconstruction. Although there was a trend toward a greater use of reconstruction with both higher levels of education and income, these independent associations were not found to be statistically significant.

Table 2. Multivariate Analysis of Significant Correlates of the Receipt of Breast Reconstruction
VariableOR95% CI
  1. OR: odds ratio; 95% CI: 95% confidence interval; AJCC: American Joint Committee on Cancer.

  2. Coefficients were weighted to account for differential selection by stage of disease, ethnicity, and nonresponse.

Study site  
 Los Angeles0.90.4–1.5
 < 50 yrs8.84.0–19.3
 50–64 yrs4.52.4–8.7
 ≥ 65 yrs1.0
 Wald chi-square test51.5P<0.001
 African American0.50.2–1.0
 Wald chi-square test3.92P=0.048
 Some high school0.40.2–1.1
 High school graduate0.60.2–0.9
 Some college0.80.5–1.4
 College graduate1.0
 Wald chi-square test6.81P=0.146
 < $20,0000.50.1–1.0
 $20,000– $69,9990.70.3–1.2
 ≥ $70,0001.0
 Wald chi-square test4.60P=0.203
Married/domestic partner  
AJCC stage  
 Wald chi-square testP <0.001 
No. of comorbidities  
 ≥ 21.0
 Wald chi-square test4.06P=0.541

To examine those factors influencing the use of breast reconstruction further, all women were asked whether the procedure was explained by their surgeon or whether they were referred to a plastic surgeon to discuss treatment options. These variables were combined in an analysis to determine the proportion of women who were aware of breast reconstruction as a surgical option. Overall, 78.2% of patients reported such discussions and no significant differences were observed based on race or study site.

Despite the high proportion of women who reported provider discussions about breast reconstruction, knowledge about the procedure was very low. Only 11.2 % of women correctly answered the 3 knowledge questions (the need for multiple surgeries with reconstruction, the impact of breast reconstruction on the detection of disease recurrence, and sensory changes in the reconstructed breast). The lack of impact of breast reconstruction on the detection of local disease recurrence, a critical factor in the decision to undergo reconstruction, was recognized by only 25.1 % of responding patients.

Women who reported that they did not undergo or were not considering breast reconstruction (n = 362) were asked a series of questions regarding barriers to the use of the procedure. The only commonly endorsed barrier to breast reconstruction for women of all races was the desire to avoid more surgery or the belief that breast reconstruction was not important (77.8% of respondents). Other barriers to reconstruction were endorsed by many fewer respondents and included: 1) the belief that the surgeon did not recommend the procedure or actively discouraged it (17.5%) and 2) not knowing about breast reconstruction (5.4%). Other barriers to reconstruction (including lack of availability, concern about the appearance of the reconstructed breast, and length of surgical recovery) were endorsed by only 3.7 % of women. Figure 1 shows that there were ethnic variations in the patient report about reasons for not undergoing or considering breast reconstruction. White women were somewhat more likely to report that reconstruction was not important or that they did not want to undergo more surgery; whereas nonwhite women were more likely to believe that the procedure was not recommended or was discouraged by their surgeon. Black patients were somewhat more likely to report that they did not know enough about breast reconstruction. Racial differences in knowledge were persistent after controlling for age, education, and stage of disease. For example, the adjusted odds ratio (OR) for getting any of the three knowledge questions correct was 0.5 for African-American women compared with white women (95% confidence interval [95% CI], 0.3–0.7; Wald chi-square test for racial groups, 16.0 [P < 0.001]). Education was also found to be positively associated with knowledge (adjusted OR of 2.6; 95% CI, 4.0–7.7 for high school graduate, some college, and college graduate vs. less than high school; Wald chi-square test, 32.7 [P < 0.001])

Figure 1.

Self-reported reasons why patients did not undergo breast reconstruction among women who did not undergo or were not considering reconstruction (n = 362). The figures were weighted to account for differential selection by stage of disease, ethnicity, and nonresponse.


The results of the current study suggest that compared with earlier reports, the use of immediate and early post-mastectomy breast reconstruction has increased. Morrow et al. noted that only 3.4% of 155,463 patients with breast carcinoma reported to the National Cancer Database between 1985–1990 underwent breast reconstruction; this figure increased to 8.3% in 1994–1995.1 In a SEER registry study from 1998, breast reconstruction was performed in 15% of patients2 whereas the results of the current study documented a 30% rate of breast reconstruction among women undergoing mastectomy between December 2001 and January 2003. Both the studies by Morrow et al.1 and Alderman et al.2 found that younger age, white race, and a diagnosis of in situ carcinoma were significant predictors of undergoing breast reconstruction. Although the overall use of breast reconstruction was higher in the current study, age, race, and stage of disease remained significant predictors of its use.

The selection of younger patients with early-stage disease for breast reconstruction is in part a reflection of sound medical judgment. A recent prospective, multiinstitutional study of complication rates after breast reconstruction reported a 52% rate of complications after immediate reconstruction with tissue expanders/implants or transverse rectus myocutaneous (TRAM) flaps, with major complications reported to occur in approximately 30% of patients.13 Awareness that postreconstruction complications may delay the administration of adjuvant systemic therapy, coupled with concerns about prolonged surgery and recovery in older patients with comorbid conditions, may motivate surgeons to selectively offer immediate or early postmastectomy breast reconstruction to patients they perceive will have a favorable risk/benefit ratio for the procedure. This is consistent with the findings from a survey of 376 consultant breast surgeons in the U.K. and Ireland, in which 88% of respondents indicated that they “usually” or “always” discussed breast reconstruction with patients undergoing mastectomy, but the majority (57%) preferred delayed reconstruction because of concerns about “interference with adjuvant therapy.”14

The results of the current study suggest that variations in patient preferences play a role in the heterogeneous patterns of treatment that were observed. In contrast to prior studies, which to our knowledge only examined the correlation between demographic factors, tumor factors, and the use of breast reconstruction, we directly surveyed patients to determine whether reconstruction was discussed as a surgical option, and the reasons why they did not to undergo the procedure. We identified a high level of awareness about breast reconstruction, and it appears that it was patient preference rather than physician advice that was the most common reason for not undergoing the procedure. However, we found variation in the treatment experience by race, with African-American women undergoing breast reconstruction less frequently than their white counterparts, even after controlling for age, level of education, and stage of disease. African-American women were somewhat more likely to endorse barriers to breast reconstruction, including not knowing about the procedure or feeling that their surgeon discouraged them from undergoing it. These findings are consistent with our previous observations regarding decision making for breast carcinoma surgical treatment, including reports by African-American women of receiving less information regarding breast-conserving surgery and fewer surgeon recommendations for the procedure.15

Investigators from the M. D. Anderson Cancer Center in Houston, Texas, examined their institutional experience with immediate breast reconstruction in 1004 women who underwent a mastectomy between January 2001 and December 2002. Breast reconstruction was performed in approximately 40.0% of white women compared with 20.2% of African-American women. The rates of reconstruction for Hispanic women were similar to those reported in white women.16 After adjustment for age, stage of disease, and body mass index, the OR for immediate breast reconstruction in African-American women was 0.34 (95% CI, 0.18–0.62) compared with white women. Alderman et al.2 also found that African-American women were significantly less likely than white women to undergo breast reconstruction (OR of 0.48). Paradoxically, in those African-American women choosing to undergo breast reconstruction, the more physically demanding autogenous tissue reconstructions were used significantly more often than in white women, who were more likely to undergo implant or expander reconstruction.2 These findings, in combination with those of the current study, suggest that less knowledge about breast reconstruction, a greater perception of barriers to the procedure, and preferences or recommendations for a type of reconstruction that significantly lengthens the time of surgery and postoperative recovery all may contribute to the lower rates of breast reconstruction performed among African-American women.


Several caveats apply to our observations. We only assessed patient perspectives regarding the decision to undergo breast reconstruction. Clinician perspectives may vary, especially with regard to oncologic concerns about reconstruction. We did not have adequate information regarding some patient, provider, and organizational attributes that may affect treatment and treatment experience. In particular, information concerning medical insurance was limited because only approximately 2% of patients in the current study sample reported no insurance and we did not have specific information regarding coverage for breast carcinoma treatments such as reconstruction. We may have failed to identify some women who underwent breast reconstruction later in their course of treatment. Indeed, approximately 14.0% of women who underwent mastectomy reported they were planning or considering breast reconstruction. Including these women in our analyses as having undergone breast reconstruction did not change the results. Of the 623 women who opted not to undergo reconstruction in the M. D. Anderson Cancer Center study, only 8% elected to undergo the procedure between 15–27 months after mastectomy,16 indicating that the results of the current study are unlikely to change significantly with further follow-up. The study was retrospective in design and patient recall of their encounters with clinicians may vary with the passage of time or because of posttreatment experiences. However, the average time from treatment to completion of the questionnaire was 7 months (range, 1-14 mos) and there was no association noted between the time from diagnosis to the completion of the questionnaire and the findings described in the current study. Finally, although our study was population based, we had to exclude all Asian women and all U.S.-born women age < 50 years with invasive carcinoma at the Los Angeles site because of participation in other studies. Therefore, the findings of the current study may not be generalizable to these groups. However, the results did not change when we restricted a secondary analysis to women age ≥ 50 years.


The current study results suggest that rates of immediate and early breast reconstruction after mastectomy continue to increase. The majority of patients are aware of breast reconstruction as a treatment option, but many choose not to undergo the procedure. The optimal proportion of mastectomy patients who should undergo reconstruction is unknown. The current study suggests that although women have a high level of awareness about breast reconstruction, their actual knowledge about what reconstruction entails and its effect on cancer outcomes is quite limited. This is particularly true among African-American patients. Considering our observation that it appears to be patient preference, not physician advice, that is the most common reason for not undergoing breast reconstruction, this lack of knowledge about the procedure is particularly noteworthy. Making sure that newly diagnosed breast carcinoma patients are aware of the availability of breast reconstruction that the procedure does not have an adverse impact on disease outcomes, and that different methods of breast reconstruction are available is a reasonable goal. Whether the lack of knowledge about breast reconstruction is because of a lack of interest in the procedure or a failure to be provided with comprehensive information should be the subject of further study.