Breast carcinoma treatment received by women with disabilities compared with women without disabilities

Authors


Abstract

BACKGROUND

Disability may make it difficult to lie flat or abduct the arm to deliver radiation therapy, imposing a high risk for radiation-induced side effects or difficulty in positioning patients for mammography. The goal of the current study was to determine the differences in treatment options experienced by women with physical disabilities compared with those without disabilities.

METHODS

Chart review of 234 women who underwent surgery for breast carcinoma between June and September 1998 in a national comprehensive cancer center was conducted. Thirty-nine of the women had physical disabilities; the remaining 195 women were without disabilities. Hierarchical logistic regression was used to determine whether women with disabilities were less likely than women without disabilities to be treated with breast-conservation surgery (BCS) or neoadjuvant chemotherapy.

RESULTS

Women with disabilities underwent BCS 38% of the time, whereas women without disabilities underwent BCS 49% of the time. Neither the presence nor absence of disability (P = 0.146) nor patient age (P = 0.747) were found to be significant predictors of BCS. However, the disease stage at the time of the surgery was reported to be a significant predictor of BCS (P = 0.007). The group of patients with disabilities received neoadjuvant chemotherapy 13% of the time, whereas women without disabilities received it 29% of the time. Disability was not found to be a significant predictor of whether a patient received neoadjuvant chemotherapy. Using hierarchical logistic regression, patient age was found to be a significant predictor of neoadjuvant chemotherapy before the disease stage was entered into the regression equation. There were no data to support the hypothesis that breast carcinoma is at an advanced stage when diagnosed in women with disabilities.

CONCLUSIONS

These findings are clinically significant in that they indicate that women with disabilities are less likely to undergo BCS and are less likely to receive neoadjuvant chemotherapy compared with women without disabilities, but the difference did not reach statistical significance. To the authors' knowledge, there are no data to support the hypothesis that disabled women are diagnosed at a more advanced stage of disease compared with women without disabilities. Cancer 2002;94:1391–6. © 2002 American Cancer Society.

DOI 10.1002/cncr.10369

Breast carcinoma is a major health concern for all women,1 particularly women with disabilities, because their physical limitations can interfere with proper screening.2, 3 Approximately 20% of all women have physical disabilities4, 5 and there has been a steady increase in the number of women aging with disabilities. This situation points to the need to study the risk factors for breast carcinoma in this population and to treat these patients as an independent group6 because therapeutic options must be designed to address their specific health needs.

The treatment of primary breast carcinoma involves various combinations of surgery, radiation therapy, systemic chemotherapy, and hormone therapy.7 There is evidence that survival rates after breast-conservation surgery (BCS) are equivalent to those reported after modified radical mastectomy (MRM) in cases of early-stage breast carcinoma.8, 9 Many authors consider BCS to be the preferred local treatment modality,10 although at the national cancer center at which the current study was conducted, offering both surgical options (BCS and MRM) is considered the standard of care.11 For women with locally advanced breast carcinoma (LABC) (TNM Stage IIB [T2N1M0: tumor > 2 cm but ≤ 5 cm in greatest dimension, metastasis to movable ipsilateral axillary lymph node(s), no distant metastasis; or T3N0M0: tumor > 5 cm in greatest dimension, no regional lymph node metastasis, no distant metastasis],12 Stage III, and regional Stage IV), the treatment of choice is neoadjuvant chemotherapy followed by surgery, postoperative chemotherapy, and radiation therapy.13, 14 However, using this “sandwich approach,” some patients with LABC can be treated with BCS.15

Disability combined with old age is a risk factor for not receiving mammograms, with a larger gap noted compared with women who are older without a disability.3 The Centers for Disease Control studied a sample of 11,399 women in which 2119 were reported to have physical limitations. Women in the age group ≥ 65 years were significantly less likely to have ever had received a mammogram: 64.7% (95% confidence interval [CI], 60.4–69%) for women with physical limitations and 73.3% (95% CI, 70.7–75.9%) for women without physical limitations. For the age group ≥ 65 years, 42.5% (95% CI, 37.7–47.3%) of the women with physical limitations were significantly less likely to have had a mammogram within the last 2 years compared with 56.5% (95% CI, 53.8–59.2%) for women without physical limitations. The incidence of early breast carcinoma increased between 1985 and 1988,11, 16 most likely because of the increased use of mammography. The mortality from breast carcinoma in the population with cerebral palsy is three times that of the general population, presumably because of delayed diagnosis or treatment.17 Anecdotally, women with physical disabilities often report that their treatment options are limited.2 These facts raised concern that the lack of mammography screenings may impact the timeliness of the diagnosis of breast carcinoma in women with disabilities.

To our knowledge, little is known regarding the effect that physical limitations have on the outcome of breast carcinoma treatment. The inability to lie flat or adequately abduct the arm, conditions that are common in women with certain disabilities, may make it difficult or impossible to deliver radiation.18, 19 A woman who cannot undergo radiation therapy is not an appropriate candidate for BCS. Data from our pilot study of 10 disabled women with breast carcinoma who were chosen at random from among women admitted to the rehabilitation floor and women who visited the breast oncology clinic during 1998 revealed that the majority (7 of 10) underwent mastectomies. At least 4 of the 10 patients had a disability factor such as limited mobility or the possibility of radiation-induced toxicity related to a disability, particularly in the presence of rheumatoid arthritis (RA) that, consequently, prevented BCS. The radiation oncologists at this national cancer center and other authors20 have irradiated patients with RA with no reportedly higher rate of radiation-induced late side effects. In our pilot study, one woman with questionable diagnosis of lupus, who was asymptomatic, elected to undergo MRM as primary treatment for her breast carcinoma. It is interesting to note that the patient was told that the possibility of radiation-induced side effects related to her disability was an additional concern in her consideration of BCS. This national cancer center21 and others22 have reported soft tissue necrosis, bone necrosis, and permanent pulmonary damage in patients with preexisting collagen vascular disease (CVD), but not in those patients who are treated with radiation when lupus is in remission or those who develop CVD after radiation therapy is complete. The findings of this pilot study indicated the need for further exploration of how the presence of a physical disability may influence therapeutic decision-making in women with breast carcinoma.

The purpose of the current study was to identify the differences in the treatment options experienced by women with disabilities compared with women without disabilities. The following hypotheses relate to the current standard of care for women with breast carcinoma and the quality of breast carcinoma care for women with disabilities:

  • 1Women with disabilities are less likely to undergo BCS.
  • 2Women with disabilities are less likely to receive neoadjuvant or preoperative chemotherapy.
  • 3Women with disabilities are more likely to be diagnosed at a more advanced stage of disease.

MATERIALS AND METHODS

We studied a historic cohort of 234 women with breast carcinoma who underwent breast carcinoma surgery at a large national cancer institute in the southern U.S. between June and September 1998. The database used in the current study was prepared by Medical Informatics (M. D. Anderson Cancer Center, Houston, TX) and the Breast Oncology Database (M. D. Anderson Cancer Center, Houston, TX) was utilized for staging. The women were divided into two groups: those with a physical disability and those without a physical disability. Disability was defined as a physical limitation that interfered with mobility, the activities of daily living, or both as noted in the diagnoses or review of systems in the patient's chart. Excluded from the current study were women with benign tumors or men who underwent breast carcinoma surgery during that period. The diagnoses of obesity, fibromyalgia, and fatigue syndromes were not considered to be diagnoses of disability.

The variables examined were age, presence or absence of a disability, type of surgery (BCS, MRM, or total masectomy [TM]), use of neoadjuvant chemotherapy, and disease stage. The treatment decision was based on medical factors and disability factors. There were four categories of medical factors:

  • 1Determinants of MRM11, 13, 14 or TM, such as large tumor size, cosmetic issues (i.e., extensive tumor excision in a small breast), social problems affecting compliance with radiation therapy (i.e., transportation to and from the clinic, being able to adhere to the frequency of the follow-up schedule), involved surgical margins, and multifocal or multicentric tumors.
  • 2Determinants for BCS, such as small, solitary tumors.14, 15
  • 3Failure of a treating physician to specify why a particular surgical method was selected.
  • 4The patient chose one type of surgery over another.10, 11

The disability factors affecting the treatment decision included the presence of a disability that limited the patient's mobility and subsequently the ability to lie flat, limitations of joint mobility that prevented abduction of the shoulder,18 or fear that the disability would increase the patient's risk for radiation-induced side effects. In some cases, the effect of the medical factors and the disability factors both were considered in determining treatment.

Hierarchal logistic regression was used to determine the incremental predictive value of disability status, age, and disease stage for predicting the use of BCS and neoadjuvant chemotherapy in two separate regression analyses. In each analysis, the three predictors were entered in order: disability status first then patient age, and finally disease stage. The final equation contains all three predictors.

RESULTS

The mean age in the group of women with disabilities was 56 years (range, 24–86 years) and was 52 years in the women without disabilities (range, 23–84 years). The disability diagnoses of these patients are listed in Table 1. The sample included a total of 234 women who underwent breast carcinoma surgery between June and September 1998, 39 of whom were classified as disabled and 195 who were classified as nondisabled.

Table 1. Disability Diagnoses with Total Number of Disabled Women and BCSs Performed
Disability diagnosesWomenBCS
  1. BCS: breast-conservation surgery.

Cerebral vascular accident20
Rheumatoid arthritis31
Polio, post-polio11
Lupus10
Mixed connective tissue21
Fractures, trauma51
Muscular weakness, foot drop11
Joint surgery21
Amputation10
Closed head injury10
Parkinson's disease10
Spinal stenosis, surgery, scoliosis50
Gout, osteoarthritis, back pain149
Total3915

BCS versus MRM/TM

Table 2 shows the statistics from the final model indicating that only disease stage was found to be a significant predictor of whether a patient underwent BCS. Neither disability status nor age was found to be a significant predictor in the previous step models.

Table 2. Variables in the Final Model: Breast-Conservation Surgery
 BSEdfSig95% CI for EXP(B)
LowerUpper
  1. B: analogous to the beta in a regression model; SE: standard error; df: degrees of freedom; Sig: significance; 95% CI: 95% confidence interval; DISA2: dichotomous variable of disabled or not disabled subject.

DISA2(1)0.5610.38610.1460.8223.733
Age0.0030.01110.7470.9831.025
Stage40.007
Stage I−0.0571.00510.9540.1326.769
Stage II1.1160.89910.2140.52517.773
Stage III0.5660.90010.5300.30210.283
Stage IV−0.6391.00710.5260.0733.797
Constant−1.3261.06910.215

Neoadjuvant Chemotherapy

Table 3 presents the statistics for the final model predicting the use of neoadjuvant chemotherapy. Patient age was found to be a significant predictor of the use of neoadjuvant chemotherapy in the second stage of this analysis (P = 0.006). However, in the following stage, entering disease stage, the previous predictive power of patient age was eroded. Table 3 indicates that only disease stage was found to be a significant predictor of the use of neoadjuvant chemotherapy in the presence of disability and patient age.

Table 3. Variables in the Final Model: Neoadjuvant Chemotherapy
 BSEdfSig95% CI for EXP(B)
LowerUpper
  1. B: analogous to the beta in a regression model; SE: standard error; df: degrees of freedom; Sig: significance; 95% CI: 95% confidence interval; DISA2: dichotomous variable of disabled or not disabled subject.

DISA2(1)1.2140.90610.1800.57119.880
Age−0.0240.01910.2000.9401.013
Stage40.000
Stage I−18.40445.45010.6860.0004.947E + 30
Stage II−13.19239.52210.7390.0008.163E + 27
Stage III−10.38739.51610.7930.0001.334E + 29
Stage IV−6.21939.52710.8750.0008.812E + 30
Constant9.49439.53010.810

Late-Stage Diagnosis

Table 4 demonstrates the distribution of the women with disabilities and those without disabilities based on disease stage at the time of diagnosis. The percentage of patients at each disease stage was approximately the same for both groups. However, to our knowledge are no data to support the hypothesis that the women with disabilities were diagnosed at a later stage of disease.

Table 4. Distribution of Women with and Women without Disabilities Based on Disease Stage
 Stage 0Stage IStage IIStage IIIStage IV
Disabled23% (9/39)41% (16/39)31% (12/39)3% (1/39)3% (1/39)
Nondisabled13% (25/195)42% (81/195)32% (62/195)13% (25/195)1% (2/195)

DISCUSSION

BCS versus MRM/TM

BCS is the treatment of choice in selected patients with early-stage breast carcinoma.8, 9, 11 Although the difference was not found to be statistically significant, the percentage of women with disabilities who received BCS was less than the percentage of women without disabilities who received this treatment. There were 4 patients who underwent >1 surgery over a period of 4 months (i.e., both BCS and MRM), but they were counted only once as having undergone BCS to facilitate the statistical analysis. In the current study, 17 of 39 patients had RA or a non-RA CVD. Disability factors were not preponderant in the decision-making process for BCS versus MRM/TM as opposed to our initial findings in the pilot study. In only 1 of 39 women was there mention of RA in addition to a medical factor that influenced the treatment decision in favor of MRM. One of the three patients with RA received BCS and radiation therapy. The remaining two patients with RA had obvious medical factors that favored MRM or the patient elected to undergo MRM/TM.

A limitation of the current study was a lack of data regarding how a patient decided between MRM/TM or BCS. Medical factors commonly accepted as determinants for MRM/TM are the size of the tumor,16 whether the tumor is multicentric or multifocal, cosmetic factors,23 contraindications to radiation therapy, poor compliance with radiation therapy protocol (i.e., frequency of sessions, transportation to clinic), or previous irradiation of the breast.11 Kotwall et al.,9 in a report of a survey of 300 Southern women, discovered a prevailing fear of radiation therapy and tumor recurrence that led women to favor mastectomy over BCS. The academic setting and the geographic region24 were found to be relevant factors in the current study that favored the decision for BCS.

Neoadjuvant versus No Neoadjuvant Chemotherapy

Neoadjuvant chemotherapy is the treatment of choice for patients with LABC.14 This approach also is being evaluated for patients with early-stage breast carcinoma.7 In a study conducted by the National Surgical Adjuvant Breast and Bowel Project (NSABP B-18), the administration of doxorubicin and cyclophosphamide chemotherapy before surgery allowed a higher rate of BCS without impairing overall survival.25 In the current study, 23% of the women with physical disabilities were diagnosed with Stage 0 disease and these patients were excluded from the data analysis concerning the use of neoadjuvant chemotherapy. Of the 41% of women with disabilities with Stage I disease, 6% (1 of 16 patients) received neoadjuvant chemotherapy whereas 1% of the women without disabilities (1 of 81 patients) received neoadjuvant chemotherapy. Approximately 6% of the women with disabilities in the current study had Stage III or Stage IV (LABC) disease, and all received neoadjuvant chemotherapy with the exception of one woman who had cardiac comorbidities that contraindicated such treatment. Stage grouping comparisons are shown on Table 3.

Of particular interest is the fact that 20% of the women with disabilities (8 of 39 women) had lymph node invasion. Of those 8 women, 38% (3 of 8 women) received neoadjuvant chemotherapy. Of the remaining five patients, one had comorbidities that contraindicated chemotherapy. One patient had a multicentric tumor. Two patients were not evaluated by the oncologist until after surgery was completed, thereby preventing a decision to use neoadjuvant chemotherapy. In one patient, it was not clear why neoadjuvant chemotherapy was not selected after it had been considered.

The oncologist did not see many of the women with Stage II tumors until after surgery. This led us to wonder whether early intervention by the medical oncologist would have changed the decision in favor of neoadjuvant chemotherapy.

Late-Stage Diagnosis

The results of the current study failed to support the hypothesis that women with disabilities are diagnosed with breast carcinoma at later stages compared with women without disabilities. The increased use of mammograms16 as a screening tool resulted in an increased diagnosis of early breast carcinoma in the 1980s. The incidence of breast carcinoma increased in the later decades but the mortality12 remained constant, perhaps from early intervention as a result of screenings16 and successful therapies.12 The physical limitations of women with disabilities may restrict them from the position required for mammography screening,2, 3 but it is not clear whether the physical limitations influence the timeliness of the diagnosis of a breast tumor or treatment decisions. The most frequent reasons given by women with disabilities for not undergoing a mammogram are that they were physically unable to get into the position required (34%), that no physician had told them they should get a mammogram (25%), and that they believed their risk for breast carcinoma was very low (23.5%).2 Our finding of no significant difference with regard to stage at diagnosis between women with disabilities and those without may be related to the small sample size of the study. Another reason might be that women with disabilities have more frequent follow-up visits with their physicians, and hence are examined more often, allowing for the early detection of tumors. The current study did not grade the severity of disability in which difficulty in positioning for a screening mammography2, 3 is expected, possibly resulting in a delayed diagnosis. This limitation was imposed by the retrospective nature of the design. An important consideration is whether the severity of limitation may impact the timeliness of the diagnosis of breast carcinoma and perhaps the treatment options presented to individuals who use a wheelchair for mobility. A larger sample with patient interviews with which to grade the severity of disability is needed in future studies.

Last, the current study did not allow us to assess the patients' contribution to the decision-making process. To our knowledge, this area is largely unexplored and should be the subject of future prospective investigations.

Although the current study had a small sample of women with disabilities (n = 39), the clinical significance of these findings are that women with disabilities are less likely to undergo BCS and are less likely to receive neoadjuvant chemotherapy compared with women without disabilities, but the difference does not appear to reach statistical significance. To our knowledge, to date there are no data to support the hypothesis that disabled women are diagnosed at a more advanced stage of disease compared with women without disabilities.

Acknowledgements

The authors would like to acknowledge the help of Vickie Williams who edited this article and prepared it for publication. They also appreciate the help of Kathy Meroney in putting the references in order and Shu Kau for allowing them the use of the Breast Oncology Database.

Ancillary