Background
Medicaid expansion under the Affordable Care Act (ACA) is one of the most transformative health policy initiatives in the past decade,1 raising the income eligibility threshold from 90% to 138% of the federal poverty level and allowing childless adults with low incomes to enroll in Medicaid. In addition, thanks to the extensive outreach efforts of the ACA, known as the “welcome mat,”2 Medicaid expansion provided people who may have been eligible for Medicaid even before expansion the opportunity to join the program.
Breast cancer is the second most common cancer in women, claiming >40,000 lives annually.3 Prior studies comparing breast cancer treatment patterns and outcomes before and after Medicaid expansion have yielded mixed results.4 However, most studies have used the National Cancer Database5,6 or state cancer registry data,7–10 both of which lack detailed data on the enrollment history of patients with cancer in Medicaid before and after cancer diagnosis. Conversely, linked cancer registry and Medicaid files offer the benefit of conducting a detailed examination of cancer outcomes across Medicaid eligibility categories, including among those who enroll in Medicaid through the Breast and Cervical Cancer Prevention and Treatment Program.11–17 The linked cancer registry and Medicaid files also allowed us to examine the heterogeneous composition of Medicaid enrollees17 and to study cancer outcomes in the context of individuals’ timing of enrollment in Medicaid relative to cancer diagnosis.14,15,17–21 Indeed, patients with cancer who enrolled in Medicaid several months prior to their cancer diagnosis (or the “stably enrolled”) experience better outcomes than those enrolling in Medicaid just as they are diagnosed with cancer (or the “emergently enrolled”).17
From a policy standpoint, there have been several competing forces at play in the period following Medicaid expansion, potentially shaping cancer care and outcomes. On the one hand, Medicaid expansion likely improved receipt of standard treatment, shortened time to treatment initiation (TTI), and improved survival given the increased access to care and use of high-value care by Medicaid enrollees post-expansion.6,22 On the other hand, the projected shortage of oncologists and radiation oncologists,23 especially following implementation of the ACA,24 may have exacerbated wait times and lengthened TTI.
The net effect of these competing forces remains unknown. In this study, we evaluate the effect of Medicaid expansion on receipt of standard treatment, TTI, and survival, hypothesizing that (1) compared patients with cancer who were emergently enrolled, the stably enrolled would have a higher probability of receiving standard treatment and shorter TTI because they may have had a greater opportunity to connect to providers when enrolled in Medicaid for longer periods of time; (2) given their higher income, patients in the ACA group post-expansion may have a greater probability than their non-ACA counterparts to receive standard treatment and shorter TTI; and (3) overall survival may have improved post-expansion, with patients with breast cancer in the ACA group experiencing more favorable survival outcomes than their post-expansion non-ACA counterparts.
Methods
This study was approved by the Ohio Department of Health, Ohio Department of Medicaid, and the Case Western Reserve University Institutional Review Board (#20191777).
Data Sources
Consistent with our previous studies,13–16 we linked the Ohio Cancer Incidence Surveillance System (OCISS) and Ohio Medicaid enrollment data using a deterministic linkage algorithm based on patients’ social security numbers, first and last names, and dates of birth.
With an estimated completeness of 98%,25 the OCISS captures incident cancer cases diagnosed among residents of Ohio, except in situ cancers of the cervix uteri and squamous and basal cell carcinoma of the skin. In addition to patient identifiers, the OCISS includes the address of residence at the time of cancer diagnosis, cancer anatomic site, the date of cancer diagnosis, cancer stage, first-course treatment modalities and the relevant dates of treatment initiation, and the date and cause of death, if applicable. Patients’ addresses were geocoded and linked with data at the census tract level from the 2013–2017 American Community Survey and the county level from the Health Resources and Services Administration to identify patients residing in Health Professional Shortage Areas (HPSAs).
From Medicaid enrollment files, we constructed the patient’s history of enrollment in Medicaid relative to their date of cancer diagnosis. We leveraged Medicaid claims data to identify comorbid conditions and treatment with trastuzumab (HER2 inhibitor targeted therapy) in patients with HER2-positive tumors. All other treatment modalities were identified from the OCISS, given that procedure codes in claims data were incomplete.
Study Population
Our study population included women aged 18 to 64 years at the time of diagnosis who were residents of Ohio, diagnosed with incident local-stage or regional-stage breast cancer in May 2011 through December 2017, and insured by Medicaid around the time of diagnosis. We excluded women who joined Medicaid through the Ohio Breast and Cervical Cancer Early Detection Program or with missing data on treatment modalities (n=2,833; see Figure S1 in the supplementary materials, available online with this article). We did not account for pregnancy status in our inclusion/exclusion criteria.
Variables of Interest
We examined the following outcomes:
- (1) Receipt of stage-appropriate standard treatment corresponding to our study period,26 as follows: For all patients, we required mastectomy or lumpectomy followed by radiation. For those diagnosed with regional-stage disease, we also required chemotherapy. Regardless of stage at diagnosis, we required patients with positive hormone receptors to receive hormonal treatment (Supplementary Table S1). Last, for patients with regional-stage disease with HER2-positive tumors, we required that they receive trastuzumab. These criteria are more stringent than those used elsewhere, in which definitive treatment was defined as surgery or neoadjuvant systemic therapy followed by surgery.27
- (2) TTI as a measure of receipt of timely treatment,27 defined as the time elapsed between the date of diagnosis and the first instance of the first treatment modality.
- (3) Survival, defined as the time elapsed between the date of diagnosis and the date of death from any cause. Patients without a date of death were censored at December 31, 2018.
Independent Variables
The main independent variable was a composite of Medicaid expansion and income eligibility threshold. Specifically, this 3-level variable included the following categories: pre-expansion for women diagnosed with breast cancer prior to 2014, post-expansion non-ACA for those diagnosed with breast cancer in 2014 or later and with an income eligibility threshold that would have qualified them for Medicaid regardless of expansion, and post-expansion ACA for women diagnosed in 2014 or later and with a higher income eligibility threshold. Identification of women in the latter 2 categories was based on the Medicaid Information Technology System Aid Categories, which incorporate the Modified Adjusted Gross Income levels. The pre-expansion category was used as the reference in our multivariable regression analyses. In a separate model, we also created a 2-level variable to analyze our outcomes of interest before and after Medicaid expansion, without accounting for income eligibility thresholds.
With regard to the timing of enrollment in Medicaid, consistent with our previous study,17 we considered women as stably enrolled if they enrolled in Medicaid at least 3 months prior to the diagnosis and remained continuously enrolled for 4 months postdiagnosis, and as emergently enrolled if they enrolled in Medicaid in the 3 months before or after cancer diagnosis.
Demographic variables included age at diagnosis, categorized as age <50, 50–54, 55–59, and 60–64 years; race/ethnicity, grouped as non-Hispanic Black, non-Hispanic White, and all other; and marital status, dichotomized as married versus not. Using the enrollment files, we identified women in the disability eligibility categories (henceforth referred to as “disabled”), and using claims data for services spanning from cancer diagnosis to 120 days postdiagnosis, we flagged the chronic conditions with which the patients presented based on Elixhauser’s list.28,29 We grouped these conditions as follows: no comorbidities, physical conditions only, mental health conditions and/or substance abuse disorders only, and physical conditions co-occurring with mental conditions and/or substance abuse disorders.
Area-level variables from the American Community Survey included median household income, educational attainment (percentage of adults with high school diploma), and percent uninsured at the census tract level, which we grouped in quartiles; HPSA (counties not HPSA, part HPSA, and whole HPSA); and county rurality to identify counties as Appalachian, non-Appalachian/non-Metropolitan, and non-Appalachian/Metropolitan.
Analytic Approach
In addition to detailed descriptive analysis, we conducted multivariable robust Poisson regression analysis for receipt of standard treatment and Cox proportional hazards models for TTI and survival. The robust Poisson regression model, which provides a robust standard error estimate and estimated covariate effects interpretable as risk ratios (RRs) is appropriate for a binary outcome as noted in the literature.30 Note that this is a loglinear model; that is, the log-transformed probability of the outcome (receipt of standard treatment) is modeled as a linear function of the covariates listed in Table 1. The proportional hazard assumption for each Cox model covariate was assessed using a score test.31 If the proportional hazard assumption was found to be violated for a given covariate, we used a stratified Cox model—in this case, by receipt of standard treatment.
Patient Characteristics of Study Population
Using these models, we conducted an analysis using a 3-level independent variable described earlier (pre-expansion, post-expansion ACA group, and post-expansion non-ACA group), which also accounted for the eligibility category post-expansion. We conducted an additional analysis to compare outcomes between pre-expansion and post-expansion periods. In both analyses, the pre-expansion group was used as the reference category.
All analyses were performed using R version 4.3.2 (R Foundation for Statistical Computing).
Results
Table 1 presents the characteristics of our study population. The number of patients in the post-expansion period was 2.2 times greater than in the pre-expansion period. However, only 35% (n=683) of those in the post-expansion period enrolled in Medicaid via the new ACA eligibility criteria under the ACA, whereas the remaining 65% (n=1,274) may have been eligible for Medicaid prior to the expansion. The lowest percentage of emergently enrolled women was in the non-ACA post-expansion group, at 13.1% (n=167), compared with 23.7% (n=208) in the pre-expansion group and 34.4% (n=235) in the ACA group. In addition, although the number of disabled women remain relatively unchanged from pre-expansion to post-expansion, their percentage dropped from 45.3% to 22.3%, with >98% of women in the post-expansion period represented in the non-ACA group.
Outcomes Between Pre-Expansion and Post-Expansion Periods
Figure 1 shows the effect of Medicaid expansion on our outcomes of interest after adjusting for patient demographics, timing of enrollment in Medicaid, tumor characteristics, and area-level attributes. Medicaid expansion was associated with a higher probability of receiving standard treatment (adjusted RR [aRR], 1.14 [95% CI, 1.06–1.22]) and shorter TTI (adjusted hazard ratio [aHR], 1.14 [95% CI, 1.04–1.24]) but not with improved overall survival (aHR, 1.00 [95% CI, 0.80–1.26]).
Effect of Medicaid expansion on receipt of standard treatment, time to treatment initiation, and survival among Medicaid beneficiaries diagnosed with local or regional breast cancer. All 3 models were adjusted by individual-level and area-level characteristics from Table 1.
Abbreviations: aHR, adjusted hazard ratio; aRR, adjusted risk ratio.
Citation: Journal of the National Comprehensive Cancer Network 22, 3; 10.6004/jnccn.2023.7104
Outcomes Accounting for Eligibility Category in the Post-Expansion Period
Table 2 presents more detailed findings from the multivariable models for receipt of standard treatment, TTI, and survival.
Multivariable Regression Analysis of Associations of Individual- and Area-Level Characteristics With Receipt of Standard Treatment, TTI, and Survival
With regard to receipt of standard treatment, we noted a 14% higher probability of receiving the standard treatment in each of the post-expansion ACA and non-ACA groups compared with women in the pre-expansion group (aRR, 1.14 [95% CI, 1.03–1.25] and 1.14 [95% CI, 1.05–1.23], respectively). Age, race/ethnicity, marital status, being disabled, emergently enrolled in Medicaid, and having multiple chronic conditions were not associated with receipt of standard treatment; however, women with regional-stage cancer at diagnosis (aRR, 0.71 [95% CI, 0.66–0.76]), positive HER2 receptor status (aRR, 0.70 [95% CI, 0.64–0.76]), and positive hormonal receptor status (aRR, 0.77 [95% CI, 0.72–0.83]) had a lower probability of receiving standard treatment than their counterparts with local-stage disease, negative HER2 receptor status, and negative hormonal status, respectively.
Table 2 also shows the factors associated with TTI from cancer diagnosis. Compared with women diagnosed in the pre-expansion period, those in the ACA and non-ACA groups diagnosed in the post-expansion period had a shorter TTI (aHR, 1.24 [95% CI, 1.11–1.39] and 1.10 [95% CI, 1.01–1.20], respectively). Other factors associated with shorter TTI included being non-Hispanic White versus non-Hispanic Black (aHR, 1.11 [95% CI, 1.01–1.22]) and being married versus nonmarried (aHR, 1.18 [95% CI, 1.08–1.29]). Compared with women with tumors with negative hormonal status, those with positive hormonal status had significantly longer TTI (aHR, 0.87 [95% CI, 0.80–0.95]).
Relative to survival outcomes, we note a greater percentage of patients who were censored in the post-expansion period compared with the pre-expansion period (90.6% [1,774/1,957] vs 77.3% [677/876]) due to the shorter follow-up period post-expansion. We observed no survival advantage in the post-expansion period, neither in the ACA nor the non-ACA group (aHR, 1.14 [95% CI, 0.79–1.64] and 0.97 [95% CI, 0.76–1.24], respectively). As expected, however, our results showed greater hazards of death in women diagnosed with regional-stage cancer compared with those diagnosed with localized disease (aHR, 2.31 [95% CI, 1.87–2.86]), among those who were disabled (aHR, 1.77 [95% CI, 1.39–2.26]), and among those with physical comorbid conditions only (aHR, 1.47 [95% CI, 1.08–1.99]) and physical comorbid conditions co-occurring with mental conditions and/or substance abuse disorders (aHR, 1.77 [95% CI, 1.31–2.38]) compared with those with no comorbid conditions.
Discussion
Our findings showed that Medicaid expansion was associated with increased receipt of standard treatment and shorter TTI but not with improved survival.
Thanks to the data granularity in the linked cancer registry and the Medicaid files, our study is among the few to examine these outcomes while accounting for individuals’ timing of enrollment in Medicaid relative to cancer diagnosis as well as by eligibility category in the post-expansion period. In addition, we were able to gain insight into the comorbidity burden borne by various subgroups of the Medicaid population, and learned that the highest percentage of women with no comorbid conditions were in the ACA group. Conversely, the highest percentage of women with physical and/or psychiatric conditions and/or substance abuse disorders was among women in the non-ACA post-expansion group (35.1%). Similarly, in the post-expansion period, patients with breast cancer who were disabled were almost exclusively represented in the non-ACA group.
Another strength of our study is its comprehensive approach to the analysis of cancer treatment in that it uses data from all treatment modalities to derive a composite measure for receipt of stage-appropriate standard treatment. Our findings are consistent with a recent study,32 which showed increased odds of patients receiving treatment if diagnosed in a state that expanded Medicaid. However, the latter study examined outcomes one treatment modality at a time and combined data from 3 states that adopted Medicaid expansion (Louisiana, Kentucky, and Arkansas), thus masking state-specific variations in outcomes. In addition, Medicaid expansion is seldom the only policy affecting outcomes, given that other local initiatives may also impact outcomes.33
Contrary to our hypothesis, we did not find a significant positive association between being stably enrolled and receipt of standard treatment of breast cancer, suggesting that Medicaid expansion may not have been able to influence referral pathways or quality of oncology care even though it had an impact on preventive screening access. As for the interpretation of why emergently enrolled women had shorter TTI, we speculate that a more advanced stage at diagnosis likely motivates patients to seek care faster than stably enrolled women. Another explanation may be that emergently enrolled women have already established care with a provider, with planned treatment initiation pending insurance arrangements. Interestingly, our findings indicated that Medicaid-insured patients with breast cancer benefited from Medicaid expansion regardless of their income eligibility threshold.
With regard to survival, we censored our data at December 31, 2018, and our data user agreement precluded us from obtaining data for a longer follow-up period. Although a greater percentage of patients were censored in the post-expansion rather than in the pre-expansion period, we do not believe that this difference invalidates our results. Nonetheless, it will be important to reexamine the impact of Medicaid expansion on breast cancer survival outcomes by including data from subsequent years.
Our study has some limitations. First, we were unable to describe the circumstances that led a patient with cancer to seek enrollment in Medicaid. Mixed-methods research is needed to gain a better insight into these findings and to elucidate the mechanisms at play. Second, except for trastuzumab, we identified all treatment modalities from cancer registry data because procedure codes in Medicaid claims data had high rates of missingness. Third, our data capture the 3 years before and 4 years after Medicaid expansion, and these trends may have changed in subsequent years,34 especially given the substantial increase in Medicaid enrollment postpandemic.35 An interrupted time series analysis might be considered as an alternative approach to the analysis. We decided against this approach because we felt the lengths of pre-expansion and post-expansion follow-ups in the present data were not long enough to determine clear trends (particularly for a survival outcome). However, this approach may be of interest in future studies. Fourth, we recognize the fact that certain treatment modalities may have been omitted by patient choice. Although we do not have the means to verify this in our data sources, changes in the representation of childless women with breast cancer in the post-expansion period implies that treatment choices in our study population may not be comparable between the pre-expansion and post-expansion periods given concerns over fertility preservation.36,37 Last, our data are specific to Ohio, and these results may not be generalizable to other states with Medicaid expansion. However, our nuanced and comprehensive analysis could only be reasonably conducted using a single state’s data due to difficulties in accessing and linking cancer registry and Medicaid data from multiple states.
Conclusions
In analyzing treatment and survival data in Ohio patients with breast cancer post–Medicaid expansion, we observed improvements in receipt of standard treatment and TTI but not in overall survival. Given the short follow-up period in our study, future studies should reexamine these outcomes with subsequent years’ data.
Acknowledgments
The authors wish to thank Mr. Long Vu for his review of our analytic approach, and Ms. Roberta Slocumb of Ohio Department of Health for her careful review of an earlier version of the manuscript.
References
- 1.↑
Levitt L. Medicaid as a potential new third rail of US politics. JAMA Health Forum 2022;3:e225492.
- 2.↑
Medicaid and CHIP Payment and Access Commission. Issue brief November 2019: changes in medicaid and CHIP enrollment. Accessed September 5, 2022. Available at: https://www.macpac.gov/publication/changes-in-medicaid-and-chip-enrollment/
- 3.↑
Giaquinto AN, Sung H, Miller KD, et al. Breast cancer statistics, 2022. CA Cancer J Clin 2022;72:524–541.
- 4.↑
Guth M, Ammula M. Building on the evidence base: studies on the effects of Medicaid expansion, February 2020 to March 2021. Accessed June 19, 2023. Available at: https://www.kff.org/medicaid/report/building-on-the-evidence-base-studies-on-the-effects-of-medicaid-expansion-february-2020-to-march-2021/
- 5.↑
Nguyen DD, Paciotti M, Marchese M, et al. Effect of Medicaid expansion on receipt of definitive treatment and time to treatment initiation by racial and ethnic minorities and at minority-serving hospitals: a patient-level and facility-level analysis of breast, colon, lung, and prostate cancer. JCO Oncol Pract 2021;17:e654–665.
- 6.↑
Takvorian SU, Oganisian A, Mamtani R, et al. Association of Medicaid expansion under the Affordable Care Act with insurance status, cancer stage, and timely treatment among patients with breast, colon, and lung cancer. JAMA Netw Open 2020;3:e1921653.
- 7.↑
Ajkay N, Bhutiani N, Huang B, et al. Early impact of Medicaid expansion and quality of breast cancer care in Kentucky. J Am Coll Surg 2018;226:498–504.
- 8.↑
Obeng-Gyasi S, Rose J, Dong W, et al. Is Medicaid expansion narrowing gaps in surgical disparities for low-income breast cancer patients? Ann Surg Oncol 2022;29:1763–1769.
- 9.↑
Bhutiani N, Hicks AC, Huang B, et al. Identifying factors influencing delays in breast cancer treatment in Kentucky following the 2014 Medicaid expansion. J Surg Oncol 2020;121:1191–1200.
- 10.↑
Chu QD, Li T, Hsieh MC, et al. Positive impact of the Patient Protection and Affordable Care Act Medicaid expansion on Louisiana women with breast cancer. Cancer 2021;127:688–699.
- 11.↑
Chien LN, Adams EK, Yang Z. Medicaid enrollment at early stage of disease: the Breast and Cervical Cancer Prevention and Treatment Act in Georgia. Inquiry 2011;48:197–208.
- 12.↑
Adams EK, Chien LN, Gabram-Mendola SG. Treatment patterns among Medicaid-eligible women with breast cancer in Georgia: are patterns different under the breast and cervical cancer prevention and treatment act? J Oncol Pract 2012;8:46–52.
- 13.↑
Koroukian SM, Bakaki PM, Golchin N, et al. Breast cancer stage and treatment among Ohio Medicaid beneficiaries with and without mental illness. J Oncol Pract 2015;11:e50–58.
- 14.↑
Koroukian SM, Bakaki PM, Htoo PT, et al. The Breast and Cervical Cancer Early Detection Program, Medicaid, and breast cancer outcomes among Ohio’s underserved women. Cancer 2017;123:3097–3106.
- 15.↑
Koroukian SM, Bakaki PM, Raghavan D. Survival disparities by Medicaid status: an analysis of 8 cancers. Cancer 2012;118:4271–4279.
- 16.↑
Koroukian SM, Bakaki PM, Schluchter M, et al. Comparing breast cancer outcomes between Medicaid and the Ohio Breast and Cervical Cancer Early Detection Program. J Oncol Pract 2015;11:478–485.
- 17.↑
Koroukian SM, Dong W, Albert JM, et al. Post-Affordable Care Act improvements in cancer stage among Ohio Medicaid beneficiaries resulted from an increase in stable coverage. Med Care 2022;60:821–830.
- 18.↑
Bradley CJ, Gardiner J, Given CW, et al. Cancer, Medicaid enrollment, and survival disparities. Cancer 2005;103:1712–1718.
- 19.↑
Bradley CJ, Given CW, Roberts C. Correlates of late stage breast cancer and death in a Medicaid-insured population. J Health Care Poor Underserved 2003;14:503–515.
- 20.↑
Koroukian SM. Assessing the effectiveness of Medicaid in breast and cervical cancer prevention. J Public Health Manag Pract 2003;9:306–314.
- 21.↑
Bradley CJ, Stevens JL, Enewold L, et al. Stage and mortality of low-income patients with cancer: evidence from SEER-Medicaid. Cancer 2021;127:229–238.
- 22.↑
Parzuchowski A, Oronce C, Guo R, et al. Evaluating the accessibility and value of U.S. ambulatory care among Medicaid expansion states and non-expansion states, 2012–2015. BMC Health Serv Res 2023;23:723.
- 23.↑
Smith BD, Haffty BG, Wilson LD, et al. The future of radiation oncology in the United States from 2010 to 2020: will supply keep pace with demand? J Clin Oncol 2010;28:5160–5165.
- 24.↑
Yang W, Williams JH, Hogan PF, et al. Projected supply of and demand for oncologists and radiation oncologists through 2025: an aging, better-insured population will result in shortage. J Oncol Pract 2014;10:39–45.
- 25.↑
Ohio Cancer Incidence Surveillance System. OCISS July 2016 quarterly newsletter. Accessed December 30, 2022. Available at: https://odh.ohio.gov/know-our-programs/ohio-cancer-incidence-surveillance-system/resources/OCISS-Quarterly-Newsletters-July-2016
- 26.↑
Gradishar WJ, Anderson BO, Balassanian R, et al. NCCN Clinical Practice Guidelines in Oncology: Breast Cancer. Version 1.2017. Accessed July 15, 2017. To view the most recent version, visit https://www.nccn.org
- 27.↑
Cone EB, Marchese M, Paciotti M, et al. Assessment of time-to-treatment initiation and survival in a cohort of patients with common cancers. JAMA Netw Open 2020;3:e2030072.
- 28.↑
Agency for Healthcare Research and Quality. Elixhauser comorbidity software refined for ICD-10-CM. Accessed July 1, 2023. Available at: https://www.hcup-us.ahrq.gov/toolssoftware/comorbidityicd10/comorbidity_icd10.jsp
- 29.↑
Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care 1998;36:8–27.
- 30.↑
Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–706.
- 31.↑
Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994;81:515–526.
- 32.↑
Laughlin AI, Li T, Yu Q, et al. Impact of Medicaid expansion on breast cancer diagnosis and treatment in southern states. J Am Coll Surg 2023;236:838–845.
- 33.↑
Eom KY, Koroukian SM, Dong W, et al. Accounting for Medicaid expansion and regional policy and programs to advance equity in cancer prevention in the United States. Cancer 2023;129:3915–3927.
- 34.↑
Lin L, Soni A, Sabik LM, et al. Early- and late-stage cancer diagnosis under 3 years of Medicaid expansion. Am J Prev Med 2021;60:104–109.
- 35.↑
Kaiser Family Foundation. Medicaid enrollment among the unemployed during the COVID-19 pandemic and beyond. Accessed December 30, 2022. Available at: https://www.kff.org/medicaid/issue-brief/medicaid-enrollment-among-the-unemployed-during-the-covid-19-pandemic-and-beyond/
- 36.↑
Sella T, Poorvu PD, Ruddy KJ, et al. Impact of fertility concerns on endocrine therapy decisions in young breast cancer survivors. Cancer 2021;127:2888–2894.
- 37.↑
Ulrich ND, Raja NS, Moravek MB. A review of fertility preservation in patients with breast cancer. Best Pract Res Clin Obstet Gynaecol 2022;82:60–68.