Impact of Preexisting Mental Illnesses on Receipt of Guideline-Consistent Breast Cancer Treatment and Health Care Utilization

Abstract

This study determined the impact of preexisting mental illnesses on guideline-consistent breast cancer treatment and breast cancer-related health care utilization. This was a retrospective, longitudinal, cohort study conducted using data from the 2006–2008 Medicaid Analytic Extract files. The target population for the study consisted of female Medicaid enrollees who were aged 18–64 years and were newly diagnosed with breast cancer in 2007. Guideline-consistent breast cancer treatment was defined according to established guidelines. Breast cancer-related health care use was reported in the form of inpatient, outpatient, and emergency room visits. Statistical analyses consisted of multivariable hierarchical regression models. A total of 2142 newly diagnosed cases of breast cancer were identified. Approximately 38% of these had a preexisting mental illness. Individuals with any preexisting mental illness were less likely to receive guideline-consistent breast cancer treatment compared to those without any preexisting mental illness (adjusted odds ratio: 0.793, 95% confidence interval [CI]: 0.646–0.973). A negative association was observed between preexisting mental illness and breast cancer-related outpatient (adjusted incident rate ratio (AIRR): 0.917, 95% CI: 0.892–0.942) and emergency room utilization (AIRR: 0.842, 95% CI: 0.709–0.999). The association between preexisting mental illnesses and breast cancer-related inpatient utilization was statistically insignificant (AIRR: 0.993, 95% CI: 0.851–1.159). The findings of this study indicate that breast cancer patients with preexisting mental illnesses experience disparities in terms of receipt of guideline-consistent breast cancer treatment and health care utilization. The results of this study highlight the need for more focused care for patients with preexisting mental illness. (Population Health Management 2015;18:449–458)

Introduction

Breast cancer is the most common cancer among women, after skin cancer. Breast cancer is also the second leading cause of cancer-related mortality in women, exceeded only by lung cancer.¹ In the year 2013, roughly 300,000 incident cases of breast cancer were expected to occur and 39,620 women were estimated to die from breast cancer in the United States.² It has been estimated that nearly 12% of the women in the United States will develop breast cancer in their lifetime.² Despite being associated with considerable morbidity and mortality, breast cancer is one of the most treatable cancers if detected early. The primary treatment of breast cancer consists of surgical removal of the tumor. Some of the aggressive types of breast cancer surgeries include radical mastectomy, modified radical mastectomy, and total mastectomy. These procedures involve removal of the whole breast that has cancer. A relatively less aggressive surgical breast cancer treatment regimen is breast-conserving surgery, a procedure performed to remove the cancer but not the breast itself. Adjuvant treatments such as radiotherapy, hormonal therapy, chemotherapy, and tissue-targeted therapies have been found to reduce the likelihood of breast cancer recurrence and are an integral part of the breast cancer treatment regimen.³

Major medical organizations, including the American Society of Clinical Oncology, the National Comprehensive Cancer Network (NCCN), and the National Institutes of Health, have issued guidelines for the treatment of breast cancer in order to inform physicians about state-of-the-art breast cancer treatments and to facilitate standard management of breast cancer patients.⁴ These guidelines are based on the breast cancer clinical research conducted over the past few decades.^5

–12 Some of the main guidelines issued by these organizations include the use of radiotherapy after breast conserving surgery in patients with Stage I and Stage II breast cancers,^3,13
–15 radiotherapy after total mastectomy in patients with tumor size greater than 5 cm and/or tumor that has spread to 4 or more axillary lymph nodes,¹⁵ chemotherapy drugs (eg, taxanes, anthracyclines, cyclophosphamide) in patients with lymph node positive breast cancer or those with tumors larger than 1 cm,^3,15 hormonal therapies such as selective estrogen receptor modulators (eg, tamoxifen) and aromatase inhibitors (eg, anastrazole, letrozole, exemestane) in patients with estrogen receptor-positive breast cancer,^3,15,16 and tissue-targeted therapies (eg, trastuzumab) in women with Human Epidermal Growth Factor Receptor 2 (HER2) positive breast cancer.^3,15

Receipt of breast cancer treatment that is compliant with the established breast cancer treatment guidelines is crucial for optimal survival in breast cancer patients.^17,18 However, despite these treatment guidelines, studies have reported that up to 55% of women with breast cancer do not receive treatment that is compliant with these guidelines.^4,19

–22

An important first step toward improving the treatment of breast cancer patients is developing an understanding of factors that affect receipt of guideline-consistent breast cancer treatment. Various patient and health care-related characteristics have been found to be associated with receipt of guideline-consistent breast cancer treatment in prior studies. Individual characteristics such as younger age,^20,23 white race,^24,25,31,32 being married,²⁵ residence in a metropolitan area,^22,24 possession of insurance,^25,32 fewer comorbidities,^23,28,29 and prior use of mammography³¹ have been associated with receipt of guideline-consistent breast cancer treatment. Hospital-related characteristics, including number of breast cancer-related surgical procedures conducted annually,^28,30 membership in multiple National Cancer Institute–funded research networks,³³ nonteaching status,²⁵ and hospital size,^34
–36 have been found to positively impact receipt of guideline-consistent breast cancer treatment. Health care access-related variables such as the number of primary care physicians and the number of radiologists in the health care service area also have been shown to be positively associated with receipt of guideline-consistent breast cancer treatment.²³

Although these studies provide useful information, limited information currently exists about the impact of preexisting mental illnesses on the receipt of guideline-consistent breast cancer treatment. Approximately 1 in 4 adults in the United States has a mental illness and nearly 50% will develop at least 1 mental illness in their lifetime.^37,38 Because of health-related issues such as poor eating and sleeping habits, lack of exercise, indulgence in smoking, alcohol, and drug abuse, and impaired immune system, individuals with mental illnesses are more likely to develop comorbid physical conditions.^39
–41 Higher incidence of breast cancer has been reported in individuals with mental illness as compared to those without any mental illness.⁴² The impairment of cognitive, emotional, and/or behavioral functioning in patients with mental illnesses could affect their receipt of mental and physical health care services.

A thorough review of the literature yielded only 1 study evaluating the impact of preexisting mental illnesses on receipt of guideline-consistent breast cancer treatment. Goodwin et al⁴³ evaluated the effect of preexisting depression on breast cancer treatment among elderly Medicare beneficiaries with breast cancer. Women with preexisting depression were 19% more likely to receive treatment that was not consistent with established standards of breast cancer care (simple mastectomy or breast-conserving surgery plus adjuvant irradiation for Stage 0, modified radical mastectomy or breast-conserving surgery with axillary dissection and adjuvant irradiation for Stage I or II, and chemotherapy for Stages III or IV) compared to women without preexisting depression.⁴³ Although the work conducted by Goodwin et al provided useful information, the authors did not consider the impact of other mental illnesses in the study. In addition, the authors only studied this relationship among elderly women.

In addition to guideline-consistent breast cancer treatment, another treatment-related attribute that is crucial for optimal health outcomes in breast cancer patients is the health care utilization during breast cancer treatment. Frequent contact with the health care system during breast cancer treatment in the form of regular physician office visits and necessary hospital stays is necessary for timely delivery of health care and prevention of unplanned hospital visits and emergency room (ER) visits. Although factors such as older age at diagnosis, higher educational level, lower quality of life, not having children, and receipt of hormonal therapy and chemotherapy have been found to be associated with health care utilization post diagnosis in breast cancer patients,^44,45 no information is currently available about the impact of preexisting mental illnesses on health care utilization post diagnosis in breast cancer patients.

The current study determined the impact of preexisting mental illnesses on the receipt of guideline-consistent breast cancer treatment and breast cancer-related health care utilization among women Medicaid enrollees diagnosed with breast cancer. The effect of mental illnesses, including mood disorders (eg, bipolar affective disorders, dysthymic disorder, major depressive disorder, adjustment reactions), psychotic disorders (eg, schizophrenia, paranoid states, nonorganic psychoses), substance abuse and dependence disorders, and other mental disorders, on receipt of guideline-consistent breast cancer treatment was evaluated in the study. In addition, the impact of preexisting mental illnesses on breast cancer-related health care utilization during the initial 12 months following diagnosis of breast cancer was determined.

Methods

Data source

The current study used the data from the 2006–2008 Medicaid analytic extract (MAX) files. The MAX files for 39 states (all states except Alaska, Hawaii, Maine, Missouri, Montana, North Dakota, Pennsylvania, South Dakota, Utah, Wisconsin, Wyoming, and the District of Columbia) were used in this study. The MAX files are created from the Medicaid Statistical Information System and are maintained by the Centers for Medicare & Medicaid Services (CMS). Information about patient demographics, eligibility, and enrollment status was available through the MAX personal summary file. Claims for inpatient services received by the patients were provided through the MAX inpatient file, whereas information about the noninstitutional services received by the Medicaid enrollees was made available through the MAX other therapy file. Details about the prescription drugs dispensed to the patients were provided through the MAX prescription drug file.^46,47 All the files were linked using a unique encrypted recipient identification number. All data were made available to the researcher in a de-identified format. Study protocol was approved by the Institutional Review Board at the University of Mississippi. Data use agreement was obtained from CMS through Research Data Assistance Center.⁴⁸

Study sample

The target population for the study consisted of women Medicaid enrollees who were: (1) continuously enrolled in Medicaid during the years 2006–2008; (2) at least 18 years of age on January 1, 2006, and younger than 65 years of age on December 31, 2008; and (3) newly diagnosed with breast cancer between January 1, 2007, and December 31, 2007. Women younger than 18 years of age were excluded from the study analysis because breast cancer is rare in adolescent women. In addition, the study excluded women aged 65 years or older because Medicare is the primary payer for these individuals. Dual eligibles (ie, women aged younger than 65 years enrolled in both Medicare and Medicaid) also were excluded from the study because of incomplete data in the MAX files.

Women newly diagnosed with breast cancer were identified using an algorithm developed by Solin et al.⁴⁹ Per this algorithm, a case of breast cancer was defined as a new case if the medical utilization data between January 1, 2007, and December 31, 2007, met 1 or more of the following 6 treatment-related criteria: (1) mastectomy (Current Procedural Terminology, 4^th edition CPT-4 codes 19180-19240); (2) partial mastectomy with lymphadenectomy (CPT-4 code 19162); (3) excision (CPT-4 code 19120, 19125, or 19126), breast biopsy (CPT-4 code 19100 or 19101), or partial mastectomy (CPT-4 code 19160) plus lymphadenectomy (CPT-4 code 38740 or 38745); (4) excision, breast biopsy, or partial mastectomy plus diagnosis of carcinoma (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 174-174.9 or 233.0); (5) excision, breast biopsy, or partial mastectomy followed by radiation therapy (CPT-4 codes 77261-77499); or (6) excision, breast biopsy, or partial mastectomy followed by chemotherapy (CPT-4 codes 96400-96549).

The date of the first record with a diagnosis of breast cancer (ICD-9-CM codes 174, 233.0, 238.3, and 239.3) for each breast cancer case was considered to be the diagnosis date. The medical records for the cases identified using Solin's algorithm were monitored to determine if there was any breast cancer diagnosis prior to the initial diagnosis. Only those cases without any prior medical record containing a diagnosis of breast cancer were considered to be incident cases. Also, only patients who did not decease before the end of the observation period (12 months after the diagnosis of breast cancer) were included in the study.

Measures

Receipt of guideline-consistent breast cancer treatment was determined based on compliance with the 2006 NCCN guidelines for the treatment of breast cancer.¹⁵ Individuals diagnosed with Stages I and II breast cancer, who received breast conserving surgery followed by radiation therapy or total mastectomy with or without radiation therapy, were considered as having received guideline-consistent breast cancer treatment. Receipt of chemotherapy for Stages III and IV breast cancer patients was considered to be guideline-consistent breast cancer treatment. The observation period for measuring guideline-consistent breast cancer treatment was 12 months post the diagnosis of breast cancer.

Breast conserving surgery was identified from the medical utilization data using CPT-4 codes 19120-19126, 19160-19162, 19180, 19182, 19200, 19220, 19240, 19260, 19271, 19272, 19290-19298, 19316-19396, and 19499, and ICD-9-CM procedure codes 8520-8525, 8534-8536, 8541-8548, and 8663. Total mastectomy was determined using CPT-4 code 19180 and ICD-9-CM procedure codes 85.41-85.44. Radiation therapy was determined based on CPT-4 codes 77261-77418, 77427-77499, and 77520-77525, ICD-9-CM procedure codes 9221-9226, and ICD-9-CM diagnosis codes V580, V661, and V671. Use of chemotherapy was identified from the medical records with associated Health Care Common Procedure Coding System codes J8520 and J8521 (capecitabine), J8530 (oral cyclophosphamide), J9070-J9097 (cyclophosphamide), J9190 (5-flurouracil), J9260 and J9250 (methotrexate), J9201 (gemcitabine), J9390 (vinorebine), J9265 (paclitaxel), J9170 (docetaxel), J9000 and J9001 (doxorubicin), J9178 (epirubicin), J9045 (carboplatin), J9060 and J9062 (cisplatin), and J9355 (trastuzumab), ICD-9-CM procedure code 9225, and ICD-9-CM diagnosis code V581, V662, and V672. In addition, records from the prescription claims data with national drug codes for the aforementioned chemotherapy drugs were used to identify chemotherapy use.⁵⁰ Guideline-consistent breast cancer treatment was considered a dichotomous variable.

Breast cancer-related health care utilization among recipients diagnosed with breast cancer was determined in the form of total number of breast cancer-related inpatient, outpatient, and ER visits during 12 months after the diagnosis of breast cancer. An inpatient visit was considered to be breast cancer related if the primary diagnosis code associated with the record was for breast cancer (ICD-9-CM code 174, 233.0, 238.3, or 239.3). Outpatient and ER visits were considered to be breast cancer related if the primary and/or secondary diagnosis code associated with the records was for breast cancer.

The 2007 and 2008 MAX inpatient and other therapy files were used to identify breast cancer-related health care utilization. Observations listed in the inpatient file were considered to be inpatient visits. The MAX inpatient file is an event-level file and therefore each observation was considered to be 1 visit. Observations with place of service codes 11 (office), 22 (outpatient hospital), 24 (ambulatory service center), 50 (federally qualified health center), 71 (state or local public health clinic), or 72 (rural health clinic), and type of service codes 08 (physicians), 10 (other practitioners), 11 (outpatient hospital), 12 (clinic), 37 (nurse practitioner services), or those with procedure codes 99201-99215, 99241-99245, 99354-99355, 99381-99429 in the other therapy file were classified as outpatient visits. The MAX other therapy file is a claim-level file, and therefore outpatient visits were defined based on the date of service. If a recipient had 1 or more observations on a particular day that conformed to the aforementioned criteria, she was considered to have had an outpatient visit on that day. Observations with a place of service code 23 in the other therapy file and those with a revenue code 450-459 or procedure codes 99281-99285 in the other therapy file or inpatient file were considered to be ER visits. Breast cancer-related inpatient, outpatient, and ER visits were considered as continuous variables.

Preexisting mental illnesses were identified based on the ICD-9-CM diagnosis codes associated with the medical records during 12 months prior to the date of diagnosis of breast cancer (see online Appendix 1, available at www.liebertpub.com/pop). Four categories of mental illnesses were considered in this study: all mood disorders, all psychotic disorders, substance abuse and dependence disorders, and other mental disorders that did not fall into the aforementioned categories.⁵¹ Recipients with ICD-9-CM codes for mental illnesses in any position in the records in the MAX inpatient and other therapy files were considered to have a mental illness.

Other variables included in the study were age at diagnosis, race, the type of reimbursement system, breast cancer stage at diagnosis, Charlson comorbidity index (CCI), location of residence, state of residence, and the number of outpatient visits in the 12 months prior to the diagnosis of breast cancer. Age at diagnosis was considered a continuous variable. Race was categorized into white, black, Hispanic or Latino, Asian, Native Hawaiian or Other Pacific Islander, American Indian or Alaskan Native, and Others (consisting of more than 1 race and unknown race). The type of reimbursement system consisted of categories of fee for service (FFS) only (recipients who were enrolled in FFS Medicaid during 2006–2008) and managed care (recipients who were enrolled in Medicaid managed care for at least 1 month during 2006–2008). The stage of breast cancer at diagnosis was identified using an algorithm developed by Yuen et al,⁵² which is based on ICD-9-CM codes. The breast cancer staging criteria stated in Yuen's algorithm have been listed in online Appendix 2, available at www.liebertpub.com/pop. The comorbidity profile of the study sample was measured using CCI (D'Hoore adaptation).⁵³ CCI was calculated based on the medical records of the recipients during the 12-month period prior to breast cancer diagnosis. Mental illnesses were excluded from the CCI calculation. The state of residence consisted of 39 categories. The location of residence was categorized as metropolitan, suburban, rural, and unknown, and was defined for each recipient based on the US Department of Agriculture rural-urban continuum codes: metropolitan location (codes 0–3), suburban location (codes 4–5), and rural location (codes 6–9).⁵⁴

The rural-urban continuum codes were determined based on the Federal Information Processing Standard code for the county of residence listed for each recipient in the 2006 MAX personal summary file. The number of outpatient visits during 12 months prior to the date of diagnosis of breast cancer was calculated based on records in the MAX other therapy files and was considered as a continuous variable.

Statistical analysis

Bivariate analyses were conducted for comparison of breast cancer patients with and without mental illnesses using chi-square tests for categorical variables and Student's t tests for continuous variables, including age at diagnosis and CCI. Wilcoxon rank-sum test was used for the comparison of number of outpatient visits during 12 months prior to breast cancer diagnosis between breast cancer patients with mental illnesses and those without mental illnesses. Means and standard deviations are reported for continuous variables and frequencies and percentages are reported for categorical variables. Hierarchical logistic regression was used to determine the impact of preexisting mental illnesses on the receipt of guideline-consistent breast cancer care. Unadjusted and adjusted analyses were performed. Odds ratios and 95% confidence intervals are reported. The impact of preexisting mental illnesses on breast cancer-related health care utilization was examined using mixed effects Poisson regression or negative binomial regression. Unadjusted and adjusted analyses were performed. The decision to use Poisson or negative binomial regression was based on the dispersion parameter (chi-square/degrees of freedom) observed after fitting the Poisson regression model. Poisson regression was used in cases wherein the dispersion parameter was ∼1, whereas negative binomial regression was used if the value of the dispersion parameter was found to be greater than 1. Separate models were fitted with breast cancer-related inpatient, outpatient, and ER visits as the dependent variables. Incident rate ratios and 95% confidence intervals are reported.

Apart from measuring the impact of any preexisting mental illness, separate models were fitted for the individual categories of preexisting mental illness (mood disorders, psychotic disorders, substance abuse and dependence disorders, and other mental disorders). All the covariates mentioned earlier were adjusted for in each of the models. The random effects of the state and county of residence were included in all the models. All analyses were performed using SAS, version 9.2 (SAS Institute Inc., Cary, NC). The SAS procedure PROC GLIMMIX was used for fitting the multivariable models.

Results

A total of 2142 incident cases of breast cancer were identified in the Medicaid population in the calendar year 2007. The sociodemographic and clinical characteristics of the study sample are presented in Table 1. The baseline demographic and clinical characteristics of breast cancer cases with and without mental disorders are also presented in Table 1. The mean age of recipients with mental illnesses was higher than those without any mental illness (P=0.047). A greater percentage of recipients with mental illnesses were white (P<0.0001) compared to those without any mental disorders, whereas the percentage of blacks (P<0.0001), Hispanics/Latinos (P<0.0001), and Asians (P<0.0001) was greater in recipients without any preexisting mental disorders compared to those with at least 1 preexisting mental disorder. A higher proportion of recipients with preexisting mental illnesses were enrolled in Medicaid managed care for at least 1 month during 2006–2008 compared to those without any preexisting mental illness (P<0.0001). A greater percentage of recipients without any preexisting mental illness resided in metropolitan areas compared to those with a preexisting mental illness (P=0.0042). No statistically significant difference was observed in the breast cancer stage at diagnosis between recipients with and without preexisting mental illnesses. The mean CCI was higher in recipients with mental illnesses than those without any mental illness (P=0.0007). The mean number of outpatient visits during 12 months prior to breast cancer diagnosis was higher in recipients with 1 or more mental illnesses than those without any mental illness (P<0.0001).

Table 1.

Demographic and Clinical Characteristics of the Study Sample

Characteristic	All (N=2,142)	No mental disorder (N=1,306)	Any mental disorder (N=806)	P value
Age at diagnosis, Mean (SD)	50.41(8.58)	50.12(8.89)	50.86(8.06)	0.047
Race, N (%)				<0.0001
White	909 (42.44)	470 (35.99)	439 (52.51)
Black	539 (25.16)	360 (27.57)	179 (21.41)
Hispanic or Latino	248 (11.58)	175 (13.40)	73 (8.73)
Asian	89 (4.15)	81 (6.20)	8 (0.96)
Native Hawaiian or other Pacific Islander	36 (1.68)	22 (1.68)	14 (1.67)
American Indian or Alaskan Native	26 (1.21)	16 (1.23)	10 (1.20)
Others	295 (13.77)	182 (13.94)	113 (13.52)
Type of reimbursement system				<0.0001
FFS only	201 (9.38)	94 (7.20)	107 (12.80)
Managed care¹	1,941 (90.62)	1,212 (92.80)	729 (87.20)
Location, N (%)				0.0042
Metropolitan	1,750 (81.70)	1,096 (83.92)	654 (81.14)
Suburban	136 (6.35)	72 (5.51)	64 (7.94)
Rural	225 (10.50)	125 (9.57)	100 (12.41)
Unknown	31 (1.45)	13 (1.00)	18 (2.23)
Stage, N (%)				0.0564
0	46 (2.15)	28 (2.14)	18 (2.15)
I	1525 (71.20)	904 (69.22)	621 (74.28)
II	378 (17.65)	239 (18.30)	139 (16.63)
III	14 (0.65)	10 (0.77)	4 (0.48)
IV	179 (8.36)	135 (9.57)	54 (6.46)
CCI, Mean (SD)	1.12 (1.68)	1.02 (1.67)	1.27 (1.69)	0.0007
Number of outpatient visits during 12 months prior to breast cancer diagnosis	12.45 (11.59)	10.41 (10.13)	15.64 (12.94)	<0.0001

CCI, Charlson comorbidity index; FFS, fee for service; SD, standard deviation.

The category managed care comprised recipients enrolled in Medicaid managed care for at least 1 month during the study period (2006–2008).

The results of hierarchical logistic regression analyses conducted to determine the impact of preexisting mental illnesses on receipt of guideline-consistent breast cancer treatment are presented in Table 2. In the multivariable analyses, recipients with 1 or more preexisting mental illnesses were found to be 20.7% less likely to receive guideline-consistent breast cancer treatment (adjusted odds ratio=0.793, 95% confidence interval [CI]=0.646–0.973) compared to those without any mental illness. In terms of individual mental illness categories, recipients with mood disorders were 24.7% less likely to receive guideline-consistent breast cancer treatment compared to those without any mental illness. Recipients with other mental disorders were 31.7% less likely to receive guideline-consistent breast cancer treatment compared to those without any mental illness. There was no statistically significant difference in the odds of receiving guideline-consistent breast cancer treatment between recipients with psychotic disorders and substance abuse and dependence disorders and those without any preexisting mental disorders.

Table 2.

Hierarchical Logistic Regression for Determination of Impact of Preexisting Mental Illnesses on Guideline-Consistent Breast Cancer Treatment

	Odds ratio
Mental illness	Unadjusted	Adjusted
No mental illness	Reference	Reference
Any mental illness	0.806 (0.673–0.967)	0.793 (0.646–0.973)
Mood disorders	0.764 (0.602–0.970)	0.753 (0.585–0.970)
Psychotic disorders	0.977 (0.667–1.431)	0.939 (0.635–1.390)
Substance abuse and dependence disorders	0.743 (0.526–1.051)	0.805 (0.559–1.159)
Other mental disorders	0.688 (0.542–0.873)	0.683 (0.531–0.879)

The results of Poisson/negative binomial regression analyses conducted to determine the impact of preexisting mental illnesses on utilization of breast cancer-related inpatient services are presented in Table 3. In the multivariable analyses, no statistically significant differences were found in terms of number of inpatient visits between recipients with and without any preexisting mental illnesses. Similar results were obtained in the analyses conducted to determine the impact of individual mental illness categories on breast cancer-related inpatient utilization. Table 4 presents the results of Poisson/negative binomial regression analyses conducted to determine the impact of preexisting mental illnesses on breast cancer-related outpatient visits. In the multivariable analyses, recipients with 1 or more preexisting mental illnesses had 8.3% fewer outpatient visits (adjusted incident rate ratio [AIRR]=0.917, 95% CI=0.892–0.942) compared to those without any mental illness. The analyses conducted to determine the impact of individual mental illness categories on breast cancer-related outpatient utilization yielded similar results. Recipients with preexisting mood disorders had 7.3% fewer outpatient visits compared to those without any mental illness. Recipients with psychotic disorders had 17.1% fewer outpatient visits compared to those without any mental illness. Recipients with substance abuse and dependence disorders had 8.5% fewer outpatient visits compared to those without any mental illness. There were 7.4% fewer outpatient visits among recipients with other mental illnesses compared to recipients without any preexisting mental illness.

Table 3.

Mixed Effects Poisson/Negative Binomial Regression for Determination of Impact of Preexisting Mental Illnesses on Breast Cancer-Related Inpatient Visits

	Incident rate ratio
Mental illness	Unadjusted	Adjusted
No mental illness	Reference	Reference
Any mental illness	0.932 (0.804–1.081)	0.993 (0.851–1.159)
Mood disorders	0.851 (0.710–1.020)	0.906 (0.749–1.097)
Psychotic disorders	0.943 (0.712–1.242)	1.001 (0.753–1.330)
Substance abuse and dependence disorders	1.034 (0.800–1.337)	1.147 (0.877–1.500)
Other mental disorders	0.990 (0.828–1.184)	1.060 (0.880–1.278)

Table 4.

Mixed Effects Poisson/Negative Binomial Regression for Determination of Impact of Preexisting Mental Illnesses on Breast Cancer-Related Outpatient Visits

	Incident rate ratio
Mental illness	Unadjusted	Adjusted
No mental illness	Reference	Reference
Any mental illness	0.882 (0.860–0.905)	0.917 (0.892–0.942)
Mood disorders	0.879 (0.852–0.908)	0.927 (0.897–0.958)
Psychotic disorders	0.792 (0.750–0.837)	0.829 (0.784–0.877)
Substance abuse and dependence disorders	0.835 (0.792–0.880)	0.915 (0.866–0.966)
Other mental disorders	0.908 (0.879–0.938)	0.926 (0.894–0.958)

The results of Poisson/negative binomial regression analyses conducted to determine the impact of preexisting mental illnesses on utilization of breast cancer-related ER services are presented in Table 5. In the multivariable analyses, recipients with any preexisting mental illnesses had 15.8% fewer ER visits (AIRR=0.842, 95% CI=0.709–0.999) compared to those without any mental illness. No statistically significant differences were observed in the number of ER visits between recipients belonging to individual mental illness categories and those without any mental illnesses.

Table 5.

Mixed Effects Poisson/Negative Binomial Regression for Determination of Impact of Preexisting Mental Illnesses on Breast Cancer-Related Emergency Room Visits

	Incident rate ratio
Mental illness	Unadjusted	Adjusted
No mental illness	Reference	Reference
Any mental illness	0.744 (0.636–0.870)	0.842 (0.709–0.999)
Mood disorders	0.769 (0.632–0.935)	0.978 (0.790–1.209)
Psychotic disorders	0.704 (0.482–1.030)	0.667 (0.431–1.031)
Substance abuse and dependence disorders	0.795 (0.566–1.119)	0.749 (0.508–1.103)
Other mental disorders	0.936 (0.787–1.113)	1.221 (0.850–1.752)

Discussion

Receipt of health care that conforms to the established breast cancer treatment guidelines is crucial for optimal health outcomes for breast cancer patients. However, it has been reported that more than 40% of the breast cancer patients do not receive the recommended health care. An understanding of factors affecting breast cancer treatment consistent with the established standards is important for planning steps toward eliminating disparities in the treatment of breast cancer patients. The current study advances the knowledge about factors affecting guideline-consistent breast cancer treatment by examining the impact of preexisting mental illnesses on the receipt of guideline-consistent breast cancer treatment among Medicaid enrollees diagnosed with breast cancer. In addition, the impact of preexisting mental illnesses on breast cancer-related health care utilization (inpatient, outpatient, and ER visits) among newly diagnosed cases of breast cancer in the Medicaid population was determined. Although a previous study had determined the impact of depression on guideline-consistent breast cancer treatment in elderly breast cancer patients,⁴³ the current study is the first to evaluate the impact of all major mental illnesses on guideline-consistent breast cancer treatment. Also, this is the first study to determine the impact of preexisting mental illnesses on health care utilization in breast cancer patients.

Recipients with a preexisting mental illness were 20.7% less likely to receive guideline-consistent breast cancer treatment compared to recipients without any preexisting mental illness. Similar results were obtained in the analyses conducted to determine the impact of individual mental illness categories on receipt of guideline-consistent breast cancer treatment. The odds ratios did not reach statistical significance for psychotic disorders and substance abuse and dependence disorders, which could be because of the low statistical power given the small sample size of these groups.

Similar to the present study, Goodwin et al found that preexisting depression was associated with 19% higher odds of non–guideline-consistent breast cancer treatment in their study of elderly breast cancer patients.⁴³ Studies of other cancers^51,55 and other disease areas^56,57 have found negative associations between preexisting mental illnesses and guideline-consistent treatment. Various patient- and provider-level characteristics can explain the negative association between the presence of preexisting mental illnesses and receipt of guideline-consistent breast cancer treatment observed in the present study. Impaired cognitive ability and poor communication skills in patients with mental illnesses could be responsible for less understanding of the treatment regimen. Social isolation and listlessness could lead to lack of motivation to undergo treatments. Disorganized thought processes could hinder the receipt of follow-up treatments.⁵¹ Patients with conditions such as paranoia, delirium, and dementia can wrongly perceive certain established treatments as life threatening and therefore not consent to receive them.^41,51,58 In terms of provider-level factors, the stigma associated with the treatment of patients with mental illnesses could cause providers to treat patients with mental illnesses differently compared to patients without mental illnesses. Also, physicians providing breast cancer treatment to patients with mental illnesses may not have the time and/or skills to provide care to these patients.^59
–61

Interesting results emerged from the multivariable regression analyses conducted to determine the impact of preexisting mental illnesses on breast cancer-related health care utilization. The association between the presence of any preexisting mental illnesses and breast cancer-related inpatient utilization was not statistically significant. Similar results were obtained in the analyses examining the impact of individual mental illness categories on breast cancer-related inpatient utilization.

However, a negative association was observed between the presence of preexisting mental illnesses and breast cancer-related outpatient visits. The results were consistent for the composite variable of any preexisting mental illnesses as well as the individual mental illness categories. Most of the breast cancer treatments including surgical treatments and systemic adjuvant therapies are provided in outpatient settings because of technological developments in breast cancer treatment. In general, only a few patients undergoing mastectomies and axillary lymph node dissections are treated in inpatient facilities.⁶² This fact also was evident in this study with outpatient visits accounting for more than 96% of the breast cancer-related health care use. Also, only ∼29.4% of the study sample had inpatient visits and the average number of inpatient visits among these recipients was 1.30 (±0.60). Considering these facts, the negative association between preexisting mental illnesses and breast cancer-related outpatient visits observed in this study is indicative of the disparities experienced by breast cancer patients with preexisting mental illnesses in terms of breast cancer-related health care utilization. It is likely that the lower number of outpatient visits among women with breast cancer with preexisting mental illnesses contributes to their lack of guideline-consistent treatment. Not having regular office visits may negatively impact their treatment.

Negative association was observed between the presence of any preexisting mental illnesses and utilization of breast cancer-related ER services. This finding was unexpected and could be explained by the possibility of lower incidence of treatment-related toxicities among those with preexisting mental illnesses related to lower use of breast cancer treatments compared to those without any preexisting mental illness. However, when the impact of any preexisting mental illnesses on all-cause ER services was examined in a secondary analysis, a significantly positive association was found.

The findings of this study have important practical implications. The negative association between preexisting mental illnesses and guideline-consistent breast cancer treatment and breast cancer-related health care utilization observed in this study emphasizes the need for more focused care of breast cancer patients with mental illnesses. Strategies that rectify the negative effects of mental illnesses such as physician counseling, health care skills training, peer-led counseling and help in accessing health care, and support from family members could be helpful in reducing the health care disparities for these individuals.⁵⁸ Provision of integrated health care by involvement of mental health professionals during the breast cancer treatment phase also could be helpful in improving breast cancer treatment for these individuals. This could be achieved by having mental health professionals visit the concerned oncologist/physician during patient visits or appointing case managers, who serve as a liaison between the specialties and coordinate the treatment of the patient.⁵⁸

The current study had a few limitations. Coding errors are possible when processing administrative claims, which could have impacted the results of this study. Individuals enrolled in both Medicare and Medicaid were not included in the study because Medicare is the primary payer for these individuals and complete information about their medical care is not contained in the MAX files. Therefore, the results obtained from this study are not representative of the entire Medicaid population in the states included in the study. The incident cases of breast cancer and cancer stage were identified using algorithms developed by Solin et al and Yuen et al. Though these algorithms have been associated with favorable measurement properties in different patient populations, they have not been validated in the Medicaid population. Mental illnesses were identified using medical claims data and ICD-9-CM diagnostic codes, which might have underestimated the true prevalence, because physicians often fail to recognize some mental illnesses, such as depression and dementia.^51,63,64 Further, this study did not consider the role of (mental health-related) medication adherence on receipt of guideline-consistent breast cancer treatment. Some of the established breast cancer treatment guidelines, including use of chemotherapy for lymph node-positive breast cancer, endocrine therapies for estrogen receptor-positive cancers, and tissue-targeted therapies for HER2-positive breast cancer, were not considered when determining guideline-consistent breast cancer treatment because of the unavailability of information in the MAX files.

This study determined the impact of preexisting mental illnesses on receipt of guideline-consistent breast cancer treatment and breast cancer-related health care utilization using multistate Medicaid data. A negative association was observed between presence of preexisting mental illnesses and guideline-consistent breast cancer treatment. The association between preexisting mental illnesses and breast cancer-related inpatient utilization was found to be statistically insignificant, whereas a negative association was observed between preexisting mental illnesses and breast cancer-related outpatient utilization. The results were found to be consistent across different mental illness categories (any mental disorder, mood disorders, psychotic disorders, substance abuse and dependence disorders, and other mental disorders) for the most part. A negative association was observed between the presence of any preexisting mental illness and breast cancer-related ER visits, whereas the results concerning the association between individual mental illness categories and breast cancer-related ER visits were statistically nonsignificant. The results of this study highlight the disparities experienced by newly diagnosed breast cancer patients with preexisting mental illnesses both in terms of receipt of guideline-consistent breast cancer treatment and breast cancer-related health care utilization. Future studies should examine the impact of preexisting mental illnesses on survival in breast cancer patients.

Footnotes

Author Disclosure Statement

Drs. Mahabaleshwarkar, Khanna, Banahan, West-Strum, Yang, and Hallam declared no conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors received no financial support for this article.

References

American Cancer Society.. What are the key statistics about breast cancer?. http://www.cancer.org/Cancer/BreastCancer/DetailedGuide/breast-cancer-key-statistics. Accessed February 20, 2014 .

American Cancer Society.. Breast cancer facts & figures 2013–2014. http://www.cancer.org/acs/groups/content/@research/documents/document/acspc-042725.pdf. Accessed January 30, 2014 .

Maughan

, Lutterbie

, Ham

. Treatment of breast cancer. Am Fam Physician. 2010, 81:1339–1346.

Iyengar

, Lund

, Lamson

, et al. Using National Quality Forum breast cancer indicators to measure quality of care for patients in an AVON comprehensive breast center. Breast J., 2010; 16:240–244.

Ragaz

, Jackson

, Le

, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med., 1997; 337:956–962.

Overgaard

, Hansen

, Overgaard

, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med., 1997; 337:949–955.

Fisher

, Anderson

, Bryant

, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med., 2002; 347:1233–1241.

Goss

, Ingle

, Martino

, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med., 2003; 349:1793–1802.

Romond

, Perez

, Bryant

, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med., 2005; 353:1673–1684.

10.

Fisher

, Wolmark

, Redmond

, Deutsch

, Fisher

. Findings from NSABP protocol no. b04: Comparison of radical mastectomy with alternative treatments. II. The clinical and biologic significance of medial‐central breast cancers. Cancer., 2006; 48:1863–1872.

11.

Coates

, Keshaviah

, Thürlimann

, et al. Five years of letrozole compared with tamoxifen as initial adjuvant therapy for postmenopausal women with endocrine-responsive early breast cancer: update of study BIG 1-98. J Clin Oncol., 2007; 25:486–492.

12.

Smith

, Procter

, Gelber

, et al. 2-year follow-up of trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer: a randomised controlled trial. Lancet., 2007; 369(9555):29–36.

13.

National Institutes of Health.. Treatment of early-stage breast cancer. http://consensus.nih.gov/1990/1990earlystagebreastcancer081html.htm. Accessed February 20, 2013 .

14.

Recht

, Edge

, Solin

, et al. Postmastectomy radiotherapy: clinical practice guidelines of the American Society of Clinical Oncology. J Clin Oncol., 2001; 19:1539–1569.

15.

National Comprehensive Cancer Network.. Breast cancer: treatment guidelines for patients. http://screening.iarc.fr/doc/Breast_VIII.pdf. Accessed February 20, 2013 .

16.

Burstein

, Prestrud

, Seidenfeld

, et al. American Society of Clinical Oncology clinical practice guideline: update on adjuvant endocrine therapy for women with hormone receptor–positive breast cancer. J Clin Oncol., 2010; 28:3784–3796.

17.

Hebert-Croteau

, Brisson

, Latreille

, Rivard

, Abdelaziz

, Martin

. Compliance with consensus recommendations for systemic therapy is associated with improved survival of women with node-negative breast cancer. J Clin Oncol., 2004; 22:3685–3693.

18.

Maskarinec

, Pagano

, Lurie

, Bantum

, Gotay

, Issell

. Factors affecting survival among women with breast cancer in Hawaii. J Womens Health (Larchmt)., 2011; 20:231–237.

19.

Bloom

, de Pouvourville

, Chhatre

, Jayadevappa

, Weinberg

. Breast cancer treatment in clinical practice compared to best evidence and practice guidelines. Br J Cancer., 2004; 90:26–30.

20.

Landercasper

, Dietrich

, Johnson

. A breast cancer review of compliance with National Comprehensive Cancer Network Breast Cancer guidelines. Am J Surg., 2006; 192:525–527.

21.

Foley

, Kimmick

, Camacho

, Levine

, Balkrishnan

, Anderson

. Survival disadvantage among Medicaid-insured breast cancer patients treated with breast conserving surgery without radiation therapy. Breast Cancer Res Treat., 2007; 101:207–214.

22.

Worthington

, Waterbor

, Funkhouser

, Falkson

, Cofield

, Fouad

. Receipt of standard breast cancer treatment by African American and white women. Int J Med Sci., 2008; 5:181–188.

23.

Ballard-Barbash

, Potosky

, Harlan

, Nayfield

, Kessler

. Factors associated with surgical and radiation therapy for early stage breast cancer in older women. J Natl Cancer Inst., 1996; 88:716–726.

24.

Haggstrom

, Quale

, Smith‐Bindman

. Differences in the quality of breast cancer care among vulnerable populations. Cancer., 2005; 104:2347–2358.

25.

Voti

, Richardson

, Reis

, Fleming

, MacKinnon

, Coebergh

. The effect of race/ethnicity and insurance in the administration of standard therapy for local breast cancer in Florida. Breast Cancer Res Treat., 2006; 95:89–95.

26.

Anderson

, Kimmick

, Camacho

, et al. Health system correlates of receipt of radiation therapy after breast-conserving surgery: a study of low-income Medicaid-enrolled women. Am J Manag Care., 2008; 14:644–652.

27.

Chagpar

, McMasters

, Scoggins

, Martin

, Thoene

, Edwards

. The use of radiation therapy after breast-conserving surgery in hormonally treated breast cancer patients is dependent on patient age, geographic region, and surgeon specialty. Am J Surg., 2008; 195:793.

28.

Rosato

, Sacerdote

, Pagano

, et al. Appropriateness of early breast cancer management in relation to patient and hospital characteristics: a population based study in Northern Italy. Breast Cancer Res Treat., 2009; 117:349–356.

29.

Jagsi

, Abrahamse

, Morrow

, et al. Patterns and correlates of adjuvant radiotherapy receipt after lumpectomy and after mastectomy for breast cancer. J Clin Oncol. 2010:28:2396–2403.

30.

Chien

, Pan

, Tsai

, et al. Radiation therapy after breast-conserving surgery: does hospital surgical volume matter? A population-based study in Taiwan. Int J Radiat Oncol Biol Phys., 2012; 82:43–50.

31.

Smith

, Shih

YCT

, Xu

, et al. Racial disparities in the use of radiotherapy after breast‐conserving surgery: A national Medicare study. Cancer., 2009; 116:734–741.

32.

Freedman

, Virgo

, He

, et al. The association of race/ethnicity, insurance status, and socioeconomic factors with breast cancer care. Cancer., 2011; 117:180–189.

33.

Laliberte

, Fennell

, Papandonatos

. The relationship of membership in research networks to compliance with treatment guidelines for early-stage breast cancer. Med Care., 2005; 43:471.

34.

Satariano

, Swanson

, Moll

. Nonclinical factors associated with surgery received for treatment of early-stage breast cancer. Am J Public Health. 1992; 82:195–198.

35.

Grilli

, Scorpiglione

, Nicolucci

, et al. Variation in use of breast surgery and characteristics of hospitals' surgical staff. Int J Qual Health Care. 1994; 6:233–238.

36.

Guadagnoli

, Weeks

, Shapiro

, Gurwitz

, Borbas

, Soumerai

. Use of breast-conserving surgery for treatment of stage I and stage II breast cancer. J Clin Oncol., 1998; 16:101–106.

37.

Kessler

, Berglund

, Demler

, Jin

, Merikangas

, Walters

. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psych., 2005; 62:593–602.

38.

Reeves

, Strine

, Pratt

, et al. Mental illness surveillance among adults in the United States. MMWR., 2011; 60(03):1–29.

39.

Kendrick

. Cardiovascular and respiratory risk factors and symptoms among general practice patients with long-term mental illness. Br J Psychiatry., 1996; 169:733–739.

40.

Osborn

. Topics in review: the poor physical health of people with mental illness. West J Med., 2001; 175:329.

41.

Mitchell

, Malone

, Doebbeling

. Quality of medical care for people with and without comorbid mental illness and substance misuse: systematic review of comparative studies. Br J Psychiatry., 2009; 194:491–499.

42.

McGinty

, Zhang

, Guallar

, et al. Cancer incidence in a sample of Maryland residents with serious mental illness. Psychiatr Serv., 2012; 63:714–717.

43.

Goodwin

, Zhang

, Ostir

. Effect of depression on diagnosis, treatment, and survival of older women with breast cancer. J Am Geriatric Soc., 2004; 52:106–111.

44.

Keyzer-Dekker

, Van Esch

, Schreurs

, et al. Health care utilization one year following the diagnosis benign breast disease or breast cancer. Breast., 2012; 21:746–750.

45.

Roorda

, Berendsen

, Groenhof

, van der Meer

, de Bock

. Increased primary healthcare utilisation among women with a history of breast cancer. Support Care Cancer., 2013; 21:941–949.

46.

Centers for Medicare & Medicaid Services.. Medicaid data sources – general information. http://www.cms.gov/Research-Statistics-Data-and-Systems/Computer-Data-and-Systems/MedicaidDataSourcesGenInfo/index.html?redirect=/MedicaidDataSourcesGenInfo/MSIS/list.asp. Accessed March 31, 2014 .

47.

Centers for Medicare & Medicaid Services.. Medicaid Analytic eXtract (MAX) general information. http://www.cms.gov/Research-Statistics-Data-and-Systems/Computer-Data-and-Systems/MedicaidDataSourcesGenInfo/MAXGeneralInformation.html. Accessed November 30, 2013 .

48.

Research Data Assistance Center.. Medicaid Analytic Extracts (MAX). http://www.resdac.org/Medicaid/file_descriptions.asp. Accessed July 25, 2013 .

49.

Solin

, Legorreta

, Schutz

, Levin

, Zatz

, Goodman

. Analysis of a claims database for the identification of patients with carcinoma of the breast. J Medical Syst., 1994; 18:23–32.

50.

Barron

, Quimbo

, Nikam

, Amonkar

. Assessing the economic burden of breast cancer in a U.S. managed care population. Breast Cancer Res Treat., 2008; 109:367–377.

51.

Baillargeon

, Kuo

, Lin

, Raji

, Singh

, Goodwin

. Effect of mental disorders on diagnosis, treatment, and survival of older adults with colon cancer. J Am Geriatr Soc., 2011; 59:1268–1273.

52.

Yuen

, Louis

, Cisbani

, et al. Using administrative data to identify and stage breast cancer cases: implications for assessing quality of care. Tumori., 2011; 97:428–435.

53.

D'Hoore

, Bouckaert

, Tilquin

. Practical considerations on the use of the Charlson comorbidity index with administrative data bases. J Clin Epidemiol. 1996; 49:1429–1433.

54.

United States Department of Agriculture.. Rural-urban continuum codes. http://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspx. Accessed July 25, 2013 .

55.

Boyd

, Benarroch-Gampel

, Sheffield

, Han

, Kuo

, Riall

. The effect of depression on stage at diagnosis, treatment, and survival in pancreatic adenocarcinoma. Surgery., 2012; 152:403–413.

56.

Druss

, Bradford

, Rosenheck

, Radford

, Krumholz

. Mental disorders and use of cardiovascular procedures after myocardial infarction. JAMA., 2000; 283:506–511.

57.

Frayne

, Halanych

, Miller

, et al. Disparities in diabetes care: impact of mental illness. Arch Intern Med., 2005; 165:2631–2638.

58.

Lawrence

, Kisely

. Inequalities in healthcare provision for people with severe mental illness. J Psychopharmacol., 2010; 24(4 suppl):61–68.

59.

Phelan

, Stradins

, Morrison

. Physical health of people with severe mental illness: can be improved if primary care and mental health professionals pay attention to it. BMJ., 2001; 322:443–444.

60.

Howard

, Barley

, Davies

, et al. Cancer diagnosis in people with severe mental illness: practical and ethical issues. Lancet Oncol., 2010; 11:797–804.

61.

Chadwick

, Street

, McAndrew

, Deacon

. Minding our own bodies: reviewing the literature regarding the perceptions of service users diagnosed with serious mental illness on barriers to accessing physical health care. Int J Ment Health Nurs., 2012; 21:211–219.

62.

Russo

, VanLandeghem

, Davis

, Elixhauser

. Hospital and Ambulatory Surgery Care for Women's Cancers. HCUP Highlight #2. AHRQ Pub. No. 06-0038. Rockville, MD: Department of Health and Human Services, Agency for Healthcare Research and Quality; 2006.

63.

Davidson

, Meltzer-Brody

. The underrecognition and undertreatment of depression: what is the breadth and depth of the problem?. J Clin Psychiatry., 1999; 60 suppl 7:4–9.

64.

Raji

, Kuo

, Freeman

, Goodwin

. Effect of dementia diagnosis on survival of older patients after a diagnosis of breast, colon, or prostate cancer: implications for cancer care. Arch Intern Med., 2008; 168:2033–2040.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB