The impact of dependent coverage expansion under the Affordable Care Act on time to breast cancer treatment among young women

Introduction Breast cancer in young women tends to be more aggressive, but timely treatment may not be always available, particularly to those without health insurance. We aim to examine whether the dependent coverage expansion under the Affordable Care Act (ACA-DCE) implemented in 2010 was associated with changes in time to treatment among women diagnosed with early stage breast cancer. Methods A total of 7,176 patients diagnosed with early stage breast cancer in 2007–2009 (pre-ACA) and 2011–2013 (post-ACA) were identified from the National Cancer Database. A quasi-experimental design difference-in-differences (DD) approach was used, with patients aged 19–25 (targeted by the policy) considered as the intervention group, and patients aged 26–34 years (not affected by the policy) as the control group. Changes in the following treatment outcomes were examined: time from diagnosis to surgery, time from surgery to adjuvant chemotherapy, and time from adjuvant chemotherapy to radiation. Results Compared with the control group of patients aged 26–34, young patients aged 19–25 experienced a statistically nonsignificant decrease of 2.7 percentage points (95% CI [-1.2, 6.5]) in the uninsured rate. This did not translate into more reduction in delays to surgery (DD = 2.7 days, 95% CI [-3.2, 8.3]), chemotherapy (DD = -1.0 days, 95% CI [-7.2, 5.2]) or radiation (DD = 5.3 days, 95% CI [-15.6, 26.3]) in the younger cohort than the older cohort. Conclusions and Relevance No significant changes in time to treatment were found among young women diagnosed with early stage breast cancer after the implementation of the ACA-DCE. Future studies examining impacts of health care policy reform on breast cancer care are warranted to include patients from low-income families and to consider effects from Medicaid expansion.


Introduction
Although breast cancer rarely occurs at young age, it still is one of the most common cancers among young adults [1,2]. Moreover, breast cancer diagnosed in women younger than 35 years tends to be more aggressive and carries a worse prognosis than in older adults [3]. While timely treatment is essential for optimized prognosis and survival of breast cancer, it is not always available to patients without adequate health insurance [4][5][6]. This may be particularly problematic for young adults, who historically had the highest uninsured rate in the US [7]. In September 2010, the dependent coverage expansion under the Affordable Care Act (ACA-DCE) went into effect, allowing young adults to be covered under their parents' health plans until they turn 26 years old. ACA-DCE has increased insurance coverage among the target population of young adults aged 19-25 years [8], as well as among newly diagnosed cancer patients of that age [2,9]. However, the impact of this policy on access to breast cancer treatment among young women is unknown. This study aimed to examine if there is any change in time to treatment after the implementation of the ACA-DCE among young women diagnosed with early stage breast cancer.

Patients
We used data from the National Cancer Database (NCDB), a nationwide hospital-based cancer registry jointly sponsored by the American Cancer Society and the American College of Surgeons, including approximately 70% of all newly diagnosed cancer cases in the U.S. [10] From the NCDB, we identified female patients aged 19-34 years old at the time of diagnosis with a first primary stage I, IIA, IIB or IIIA-T3N1M0 breast cancer in 2007-2009 and in 2011-2013.
The year 2010 was excluded as a washout/phase-in period. A quasi-experimental design difference-in-differences (DD) approach was used, with patients aged 19-25 (targeted by the policy) considered as the "intervention" group, and patients aged 26-34 years (not affected by the policy) as the "control" group. Only early stage breast cancer patients were included in the study to focus on patients who received breast surgery as their first treatment, sometimes followed by adjuvant radiation and/or systemic therapy.
We excluded patients receiving no surgery, with autopsy pathology only, with local tumor destruction, or with surgery data missing (n = 325); any patient whose surgery date or diagnosis date was missing (n = 185), or with a diagnosis date after the surgery date (n = 398); patients whose radiation or systematic therapy date was missing if they received radiation or systematic therapy (n = 650); and patients who received neoadjuvant therapy (n = 3,619). Finally, a total of 7,176 female breast cancer patients were available for the analyses.

Outcomes
Our outcomes of interest include insurance coverage, time from diagnosis to the most definitive breast surgery, time from surgery to chemotherapy if adjuvant chemotherapy was received (N = 5556), receipt of adjuvant radiation therapy if lumpectomy was the definitive surgery (N = 2527), and time from adjuvant chemotherapy to radiation if lumpectomy and both adjuvant treatments were received (N = 1710). Those who received mastectomy were excluded in the radiation analyses because changing plastic surgery practices may have impacted the time to radiation substantially. Because previous research showed that a delay of more than 2 months from diagnosis to initial treatment [11] and a delay of 2-3 months in adjuvant therapy [5,[12][13][14] were associated with worse outcomes among breast cancer patients, we examined the proportion of patients who received surgery more than 2 months after the diagnosis and proportions of patients who received adjuvant therapy more than 2 months and 3 months.

Statistical analysis
We used a difference-in-difference (DD) approach to evaluate the impact of the ACA-DCE on insurance coverage and treatment, where changes from before to after the ACA-DCE (2007-2009 vs. 2011-2013) were calculated for the intervention group of patients aged 19-25 and for the control group of patients aged 26-34. Crude uninsured rate was calculated. DD estimates and p-values for treatment outcomes were calculated using multivariable linear probability models adjusted for age, race/ethnicity, zip-code level education (percentage of residents in N=12353 female aged 19-34 years who were diagnosed with breast cancer in 2007-2009 and 2011-2013, and in stage I, IIA, IIB, IIIA-T3 N1 M0   N=12028   N=11843   N=11445 325 did not receive surgery 185 whose surgery date or diagnosis date was not available 398 whose surgery date was before diagnosis date 650 who received non-surgery therapy but the nonsurgery therapy date was not available N=10795 patient's zip code without a high school diploma), region, stage, comorbidity score [15], and facility type. Surgery type (lumpectomy or mastectomy) and reconstruction status were also controlled in the analyses of time to surgery and time to adjuvant chemotherapy. All analyses were conducted using SAS 9.4 (SAS Institute, Cary NC). Significance level was set at 0.05, and all statistical tests were two-sided.

Results
The study sample was composed of 6.0% patients in the intervention group and 94.0% in the control group. The majority of patients were non-Hispanic white (61.8%), privately-insured (78.4%), diagnosed at stage II (59.7%), and without comorbidity at the time of diagnosis (94.1%) (  Table 2). Similarly, we did not find a difference between the age groups in the change of the proportion of patients who received surgery later than 2 months after diagnosis (DD = -0.5 ppt, 95% CI [-7.

Discussion
We examined changes in insurance coverage and receipt of treatment among young women diagnosed with early stage breast cancer following the ACA dependent expansion insurance coverage using the NCDB from 2007-2013. We found a nonsignificant net decrease of 2.7 ppt in uninsured rate among patients aged 19-25 years relative to those patients aged 26-34 year following the ACA, which was comparable to the findings of two previous studies on young adult cancer patients using population-based cancer registry data [2,9], where a net decrease in uninsured rate of 3.1 ppt and 2.0 ppt rate were found respectively. We did not find any significant differences in either age group or between the age groups in pre-to post-ACA changes in receipt of treatment or treatment delays including time from diagnosis to surgery, time from surgery to adjuvant chemotherapy, receipt of adjuvant radiation after lumpectomy, and time from adjuvant chemotherapy to radiation.
Although NCDB captures 70% of new cancer cases nationwide each year, breast cancer is rare among young adults, especially among the ACA-DCE extended parental insurance eligibility-targeted population of individuals with age 19-25 years old. Thus, a relatively small sample size limits the conclusiveness of our results. Also, the majority of beneficiaries of the extended parental insurance eligibility clause in the ACA-DCE, whose parents are covered by employer-sponsored or self-purchased private insurance, were likely from families that were relatively well-off financially [16]. For such patients, insurance coverage may not be the major barrier to access to treatment. Instead, increased use of different imaging modalities and delays introduced by genetic testing and frequent second opinions [17] might be more common Limitations of our study include relatively short follow-up time since the ACA-DCE, which is especially important given that there is inevitably a time lag between a policy implementation and ultimate impact on care; limited generalizability given that our sample was from Commission-on-Cancer accredited hospitals instead of a population-based sample; lack of information on other factors that may have impacted treatment delays over this period, such as increasing adoption of genomic assays and Oncotype Dx test; and unavailability of parents' socioeconomic status to control for in the analyses.

Conclusions
In summary, this study found no statistically significant changes in time to breast cancer treatment among women 19-25 years old compared to slightly older women after the implementation of the ACA-DCE. Moving forward, studies examining the impact of the ACA on breast cancer care are warranted to include more patients from low-income families and to take Medicaid expansion into account.