Has the Affordable Care Act Been Associated with Increased Insurance Coverage and Early-stage Diagnoses of Bone and Soft-tissue Sarcomas in Adults?

Abstract

Background

Treatment of bone and soft-tissue sarcomas can be costly, and therefore, it is not surprising that insurance status of patients is a prognostic factor in determining overall survival. Furthermore, uninsured individuals with suspected bone and/or soft-tissue masses routinely encounter difficulty in obtaining access to basic healthcare (such as office visits, radiology scans), and therefore are more likely to be diagnosed with later stages at presentation. The Patient Protection and Affordable Care Act (ACA) mandate of 2010 aimed to increase access to care for uninsured individuals by launching initiatives, such as expanding Medicaid eligibility, subsidizing private insurance, and developing statewide mandates requiring individuals to have a prescribed minimum level of health insurance. Although prior reports have demonstrated that the ACA increased both coverage and the proportion of early-stage diagnoses among patients with common cancers (including breast, colon, prostate, and lung), it is unknown whether similar improvements have occurred for patients with bone and soft-tissue sarcomas. Understanding changes in insurance coverages and stage at diagnosis of patients with bone and soft-tissue sarcomas would be paramount in establishing policies that will ensure orthopaedic cancer care is made equitable and accessible to all.

Questions/purposes

(1) Has the introduction of the ACA been associated with changes in insurance coverage for adult patients with newly diagnosed bone and soft-tissue sarcomas? (2) Did the introduction of health reforms under the ACA lead to an increased proportion of sarcoma diagnoses occurring at earlier disease stages?

Methods

The 2007 to 2015 Surveillance, Epidemiology and End Results database was queried using International Classification of Diseases for Oncology codes for primary malignant bone tumors of the upper and lower extremity (C40.0 to C40.3), unspecified or other overlapping bone, articular cartilage, and joint and/or ribs, sternum, or clavicle (C40.8 to C40.9, C41.3, and C41.8 to C41.9), vertebral column (C41.2), pelvis (C41.4, C41.8, and C41.9), and soft-tissue sarcomas of the upper or lower extremity and/or pelvis (C49.1, C49.2, and C49.5). A total of 15,287 patients with newly diagnosed cancers were included, of which 3647 (24%) were malignant bone tumors and 11,640 (76%) were soft-tissue sarcomas. The study sample was divided into three cohorts according to specified time periods: pre-ACA from 2007 to 2010 (6537 patients), pre-Medicaid expansion from 2011 to 2013 (5076 patients), and post-Medicaid expansion from 2014 to 2015 (3674 patients). The Pearson chi square tests were used to assess for changes in the proportion of Medicaid and uninsured patients across the specified time periods: pre-ACA, pre-expansion and post-expansion. A differences-in-differences analysis was also performed to assess changes in insurance coverage for Medicaid and uninsured patients among states that chose to expand Medicaid coverage in 2014 under the ACA’s provision versus those who opted out of Medicaid expansion. Since the database switched to using the American Joint Commission on Cancer (AJCC) 7th edition staging system in 2010, linear regression using data only from 2010 to 2015 was performed that assessed changes in cancer stage at diagnosis from 2010 to 2015 alone. After stratifying by cancer type (bone or soft-tissue sarcoma), Pearson chi square tests were used to assess for changes in the proportion of patients who were diagnosed with early, late, and unknown stage at presentation before Medicaid expansion (2011-2013) and after Medicaid expansion (2014-2015) among states that chose to expand versus those who did not.

Results

After stratifying by time cohorts: pre-ACA (2007 to 2010), pre-expansion (2011 to 2013) and post-expansion (2014 to 2015), we observed that the most dramatic changes occurred after Medicaid eligibility was expanded (2014 onwards), with Medicaid proportions increasing from 12% (pre-expansion, 2011 to 2013) to 14% (post-expansion, 2014 to 2015) (p < 0.001) and uninsured proportions decreasing from 5% (pre-expansion, 2011 to 2013) to 3% (post-expansion, 2014 to 2015) (p < 0.001). A differences-in-differences analysis that assessed the effect of Medicaid expansion showed that expanded states had an increase in the proportion of Medicaid patients compared with non-expanded states, (3.6% [95% confidence interval 0.4 to 6.8]; p = 0.03) from 2014 onwards. For the entire study sample, the proportion of early-stage diagnoses (I/II) increased from 56% (939 of 1667) in 2010 to 62% (1137 of 1840) in 2015 (p = 0.003). Similarly, the proportion of unknown stage diagnoses decreased from 11% (188 of 1667) in 2010 to 7% (128 of 1840) in 2015 (p = 0.002). There was no change in proportion of late-stage diagnoses (III/IV) from 32% (540 of 1667) in 2010 to 31% (575 of 1840) in 2015 (p = 0.13).

Conclusion

Access to cancer care for patients with primary bone or soft-tissue sarcomas improved after the ACA was introduced, as evidenced by a decrease in the proportion of uninsured patients and corresponding increase in Medicaid coverage. Improvements in coverage were most significant among states that adopted the Medicaid expansion of 2014. Furthermore, we observed an increasing proportion of early-stage diagnoses after the ACA was implemented. The findings support the preservation of the ACA to ensure cancer care is equitable and accessible to all vulnerable patient populations.

Level of Evidence

Level III, therapeutic study.

Introduction

Hailed as the largest healthcare reform since the creation of Medicaid and Medicare in 1965, the Patient Protection and Affordable Care Act (ACA) of 2010 was a major step toward reducing rising healthcare costs and enhancing the quality of care in the United States [18, 20, 27, 29]. Although the ACA has multiple provisions and mandates affecting various aspects of healthcare, the three key mechanisms through which it enhanced cancer care delivery are increased coverage through various insurance reforms [1, 13, 19], introduction of alternative value-based payment models in oncology care [8, 22], and provision of support and increased eligibility for cancer research [28]. Increasing insurance coverage was the primary policy target through which the ACA was successful in improving access to care for individuals in the United States.

Most bone and soft-tissue sarcomas are assessed and treated by a multidisciplinary team consisting of surgery and radiation and/or chemotherapy, with the costs of care often exceeding USD 100,000 per patient [34]. Given the costs, it is not surprising that a patient’s insurance status is a major prognostic factor for overall survival [15, 21, 31]. In addition, a past report has shown that because uninsured patients often have difficulty gaining access to primary care or diagnostic services, uninsured patients with cancer are more likely to have a diagnosis of late-stage disease at presentation than are insured patients [33]. Although recent evidence has shown the ACA effectively increased insurance coverage and early-stage diagnoses for individuals with colorectal [9, 24], breast [12, 14], gynecologic [23], lung [24], and head and neck cancers [4], no study that we know of has evaluated the same for patients with bone and soft-tissue sarcomas. Despite their high morbidity, if diagnosed early, most bone and soft-tissue sarcomas have a favorable outcome in terms of survival/cure. Therefore, understanding the impact of the ACA on increasing early-stage diagnoses for these cancers would facilitate discussion on the long-term cost-effectiveness of this healthcare reform. Furthermore, given the increasing uncertainty around the future of the ACA, there is a continuous need for evidence that evaluates the healthcare reform’s impact on ensuring that orthopaedic cancer care is made accessible for all.

In light of these observations, we used a national cancer surveillance dataset to answer our primary research questions: (1) Has the introduction of ACA been associated with changes in insurance coverage for adult patients with newly diagnosed bone and soft-tissue sarcomas? (2) Did the introduction of health reforms under the ACA lead to an increased proportion of sarcoma diagnoses occurring at earlier disease stages?

Patients and Methods

Database and Patient Selection

We performed a retrospective review using registry data from the Surveillance, Epidemiology, and End Results (SEER) database. The SEER database is a national cancer registry that records incidence and cancer follow-up data from more than 18 tumor registries nationally [7]. All data recorded in the database are presented in a deidentified and Health Insurance Portability and Accounting Act-compliant format; therefore, the study was exempt from institutional review board approval. The 2007 to 2015 SEER data files were queried using International Classification of Diseases for Oncology, 3rd Edition codes for primary malignant bone tumors of the upper and lower extremity (C40.0 to C40.3), unspecified or other overlapping bone, articular cartilage, joint, ribs, sternum, or clavicle (C40.8 to C40.9, C41.3, and C41.8 to C41.9), vertebral column (C41.2: bone only), pelvis (C41.4, C41.8, and C41.9) and soft-tissue sarcomas of the upper or lower extremity and/or pelvis (C49.1, C49.2, and C49.5). At the time the study was conducted, our agreement with the SEER registry was for data until 2015. Because most of the ACA’s provisions targeted adult patients and the low number of uninsured children with bone and soft-tissue sarcomas (n = 42) in the SEER database precluded us from performing an adequately powered analysis of differences in insurance coverage over time for the pediatric population, only individuals aged 18 years or older were included in the study. The SEER database codes insurance status into the following groups: insured (private, Medicare, and Medicare advantage), Medicaid, uninsured, and unknown insurance status. Patients 65 years or older with an unknown insurance status were deemed Medicare-eligible and were recoded as falling into the “insured” category. Patients with missing and/or unknown insurance status (n = 184) were removed from the study cohort. Given that this only represents around 1% of the entire cohort, excluding these patients does not result in any major bias.

Baseline Clinical Characteristics of the Patient Population Pre-ACA and Post-ACA

After applying the inclusion and exclusion criteria, we included 15,287 adults with a diagnosis of a primary malignant bone (24%, n = 3647) or soft-tissue sarcoma (76%, n = 11,640) in the study (Table 1). Baseline clinical characteristics retrieved from the database included age (presented as the mean ± SD), gender, race (white, black, other, and unknown), type of cancer (bone or joints versus soft-tissue sarcoma), primary site (upper limb, lower limb, pelvis, vertebral column, and other or unspecified), insurance status (insured, Medicaid, and uninsured), and American Joint Commission on Cancer (AJCC) stage at the time of diagnosis (categorized into early Stage [I and II], late stage [III and IV], and unknown stage). Considering the ACA’s expansion of Medicaid eligibility, which came into effect in 2014, we stratified the cohorts into specified time periods: pre-ACA (2007 to 2010), pre-Medicaid expansion (2011 to 2013), and post-Medicaid expansion (2014 to 2015). Based on these time periods, there were a total of 6537 patients in pre-ACA, 5076 patients in the pre-expansion and 3674 in post-expansion cohorts. There was a slightly greater proportion of men in the pre-expansion (57%; 2875 of 5076) and post-expansion (56%; 2054 of 3674) cohorts compared with pre-ACA (54%; 3528 of 6537); otherwise, there were no appreciable differences in the patient populations across these three time periods.

Table 1.

Baseline demographic and clinical characteristics of the study population pre-ACA, pre-Medicaid expansion, and post-Medicaid expansion (n = 15,287)

Category		Post-ACA
	Pre-ACA (2007 to 2010)	Pre-Medicaid expansion (2011 to 2013)	Post-Medicaid expansion (2014 to 2015)	p value
Number of new diagnoses	43% (6537)	33% (5076)	24% (3674)
Age in years, median (interquartile range)	56 (42 to 70)	57 (42 to 69)	57 (42 to 69)	0.441a
Gender				0.01
Women	46% (3009)	43% (2201)	44% (1620)
Men	54% (3528)	57% (2875)	56% (2054)
Race				0.007
White	80% (5230)	80% (4080)	78% (2870)
Black	11% (749)	11% (551)	12% (428)
Other	8% (505)	8% (410)	9% (327)
Unknown	1% (53)	1% (35)	1% (49)
Type of cancer				0.69
Bones and joints	24% (1582)	24% (1196)	24% (869)
Soft-tissue sarcoma	76% (4955)	76% (3880)	76% (2805)
Primary site location				0.39
Upper limb	20% (1276)	20% (989)	21% (756)
Lower limb	55% (3618)	56% (2849)	56% (2053)
Pelvis	19% (1264)	18% (932)	18% (662)
Vertebral column (bone only)	2% (163)	2% (115)	2% (75)
Other/unspecified	3% (216)	4% (191)	4% (128)
Insurance status				< 0.001
Medicaid	11% (701)	12% (583)	14% (512)
Private/Medicare/other government	85% (5553)	84% (4262)	83% (3056)
Uninsured	4% (283)	5% (231)	3% (106)
AJCC stage				< 0.001
I	28% (1809)	38% (1915)	38% (1404)
II	16% (1069)	20% (1022	23% (860)
III	14% (938)	21% (1059)	20% (753)
IV	13% (878)	12% (620)	10% (378)
Unknown/not applicable	28% (1843)	9% (460)	8% (279)

^aKruskall-Wallis test; ACA = Affordable Care Act; AJCC = American Joint Commission on Cancer.

Primary and Secondary Study Outcomes

Our primary study outcome was to assess changes in proportion of insured, Medicaid and uninsured patients from 2007 to 2015, across the specified time periods: pre-ACA (2007 to 2010), pre-Medicaid expansion (2011 to 2013), and post-Medicaid expansion (2014 to 2015). We also aimed to assess the effect of Medicaid expansion of 2014 in changing insurance coverages for states that chose to opt into the expansion versus states that did not expand Medicaid coverage.

Our secondary study outcome was to evaluate whether the ACA was associated with changes in early, late, and unknown stage diagnoses for all patients presenting with bone and soft-tissue sarcomas from 2010 to 2015. We also aimed to assess whether Medicaid expansion of 2014 was associated with any changes in the proportion of early, late, and unknown stage diagnoses for all bone and soft-tissue sarcomas at both expanded and non-expanded states.

Statistical Analysis

Pearson chi square tests were used to assess changes in the proportion of insured, Medicaid, and uninsured patients across three specified time periods: pre-ACA (2007 to 2010), pre-Medicaid expansion (2011 to 2013), and post-Medicaid expansion (2014 to 2015). We performed a differences-in-differences analysis to assess whether the Medicaid expansion of 2014 was associated with changes in the proportion of Medicaid and uninsured patients at expanded versus non-expanded states. Difference-in-difference analyses are effective research tools for assessing policy changes, as they mimic an experimental study design by assessing the effect of an intervention (such as Medicaid expansion in the latter example) that was implemented at a certain time point (like 2014), across treatment (expanded states) and control (non-expanded states) groups across time. Linear regression analyses were used to assess changes in the difference of Medicaid and uninsured individuals across time, at both expanded and non-expanded states. Because the SEER database converted from the AJCC, 6th Edition, to the AJCC, 7th Edition, in 2010, we only analyzed stage shifts from 2010 to 2015 (using the 7th Edition for staging only) alone. This was important because under the AJCC, 6th Edition, the presence of Mx (metastases could not be assessed) resulted in tumors being classified as “unknown.” In the 7th Edition, however, Mx is treated as M0 (no metastases), and other staging information is used to carefully stage these patients’ tumors [5]. Linear regression analyses were used to assess changes in the proportion of patients with early-stage (I or II), late-stage (III or IV), and unknown stage at diagnoses from 2010 to 2015. We used an additional sensitivity analysis, using Pearson chi square tests to assess whether there were any changes in the proportion of patients with early, late and/or unknown stage diagnoses of bone and soft-tissue sarcomas, pre-expansion (2011 to 2013) and post-expansion (2014 to 2015) across both expanded and non-expanded states. For all statistical purposes, a p value of less than 0.05 was considered significant. All statistical analyses were performed using SPSS version 22 (IBM Corp, Armonk, NY, USA) and Microsoft Excel (Microsoft Office, Redmond, WA, USA).

Results

Insurance Coverage Before and After the ACA

After stratifying by time cohorts: pre-ACA (2007 to 2010), pre-expansion (2011 to 2013) and post-expansion (2014 to 2015), we observed that most-dramatic changes occurred after Medicaid eligibility was expanded (2014 onwards). The proportion of patients with Medicaid insurance rose from 11% (pre-ACA, 2007 to 2010), to 11.5% pre-expansion (2011 to 2013) and 14% post-expansion (2014 to 2015) (p < 0.001). Similarly, the proportion of uninsured patients was largely unchanged from 4% pre-ACA (2007 to 2010) to 5% pre-expansion (2011 to 2013), but decreased to 3% (post-expansion, 2014 to 2015) (p < 0.001) (Table 1). A differences-in-differences analysis that assessed the effect of Medicaid expansion showed that expanded states had an increase in the proportion of Medicaid patients from 2014 onward compared with non-expanded states (3.6% [95% CI 0.4 to 6.8]; p = 0.023). However, Medicaid expansion was not associated with changes in proportion of uninsured patients between expanded and non-expanded states (-0.7% [95% CI -1.4 to 2.7]; p = 0.05) from 2014 onwards, likely indicating that other reforms of the ACA may have contributed to the slight decrease in the proportion of uninsured patients between pre-expansion and post-expansion time periods. These findings were corroborated by the fact that there was an increasing trend in the difference of the proportion of Medicaid and uninsured individuals in expanded states (p = 0.002) over time (Fig. 1A). However, no change was noted for the difference in the proportion of Medicaid and uninsured individuals in non-expansion states (p = 0.93) over time (Fig. 1B).

Fig. 1.

A-B These graphs show trends in insurance coverage for all cancer types for (A) expansion and (B) non-expansion states.

After stratifying for cancer type, we noted declines in the proportions of uninsured patients and increases in Medicaid coverage among states that expanded Medicaid, for both bone and soft-tissue sarcomas (Table 2). Among expanded states, the proportion of Medicaid patients with bone tumors increased from 14% (185 of 1290) pre-ACA, 2007 to 2010, to 16% (156 of 955) pre-expansion, 2011 to 2013, and 19% (134 of 701) post-expansion, 2014 to 2015 (p = 0.02). Similarly, among expanded states, the proportion of uninsured patients with bone tumors was unchanged from 4% (52 of 1290) pre-ACA, to 4% (40 of 955) pre-expansion, but decreased to 2% (17 of 701) in post-expansion (p = 0.02). Among expanded states, the proportion of Medicaid patients with soft-tissue sarcomas increased from 10% (390 of 3957) pre-ACA (2007-2010) to 11% (353 of 3192) pre-expansion (2011-2013) and 14% (315 of 2270) post-expansion (2014-2015; p < 0.001). Similarly, the proportion of uninsured patients with soft-tissue sarcomas changed from 3% (132 of 3957) pre-ACA to 4% (116 of 3192) pre-expansion and 2% (41 of 2270) post-expansion (p < 0.001). Cancer Stage Shifts Under the ACA

Table 2.

Differences in insurance coverage proportions among expanded and non-expanded states, after stratifying for the type of cancer

Category	Type	Pre-ACA (2007 to 2010)	Pre-Medicaid expansion (2011 to 2013)	Post-Medicaid expansion (2014 to 2015)	p value
Expanded states	Bone tumors	n = 1290	n = 955	n = 701
	Insured	82% (1053)	79% (759)	79% (550)	0.02
	Medicaid	14% (185)	16% (156)	19% (134)
	Uninsured	4% (52)	4% (40)	2% (17)
Non-expanded states	Bone tumors	n = 292	n = 241	n = 168
	Insured	80% (232)	79% (191)	80% (135)	0.17
	Medicaid	13% (39)	13% (31)	10% (17)
	Uninsured	7% (21)	8% (19)	10% (16)
Expanded states	Soft-tissue sarcomas	n = 3957	n = 3192	n = 2270
	Insured	87% (3435)	85% (2723)	84% (1914)	< 0.001
	Medicaid	10% (390)	11% (353)	14% (315)
	Uninsured	3% (132)	4% (116)	2% (41)
Non-expanded states	Soft-tissue sarcomas	n = 998	n = 688	n = 535
	Insured	84% (833)	86% (589)	85% (457)	0.20
	Medicaid	9% (87)	6% (43)	9% (46)
	Uninsured	8% (78)	8% (56)	6% (32)

ACA = Affordable Care Act.

For the entire study sample, the proportion of early-stage diagnoses (I/II) increased from 56% (939 of 1667) in 2010 to 62% (1137 of 1840) in 2015 (p = 0.003). Similarly, the proportion of unknown stage diagnoses decreased from 11% (188 of 1667) in 2010 to 7% (128 of 1840) in 2015 (p = 0.002). There was no change in proportion of late-stage diagnoses (III/IV) from 32% (540 of 1667) in 2010 to 31% (575 of 1840) in 2015 (p = 0.13) (Fig. 2). After stratifying by sarcoma type, we observed that after Medicaid expansion, the proportion of early-stage primary bone sarcomas changed from 67% (800 of 1196) pre-expansion (2011-2013) to 76% (656 of 869) post-expansion (2014-2015; p < 0.001), whereas the proportions of late-stage cancers changed from 18% (218 of 1196) pre-expansion (2011-2013) to 14% (121 of 869) post-expansion (2014-2015; p < 0.001), and unknown-stage disease changed from 15% (178 of 1196) pre-expansion to 11% (92 of 869) post-expansion (Table 3). For primary bone sarcoma, the increases in early-stage and decrease in late-stage and unknown-stage diagnoses were seen in states that expanded Medicaid and those who did not (Table 3). However, for soft-tissue sarcomas, we did not observe a difference in the association between Medicaid expansion and early-stage, late-stage, or unknown-stage diagnoses. This was also true for stage shifts for soft-tissue sarcomas observed in both expanded and non-expanded states.

Fig. 2.

This graph shows trends in early-stage (I and II), late-stage (III and IV), and unknown-stage tumors at the time of diagnosis from 2007 to 2015.

Table 3.

Differences in stage shifts among expanded and non-expanded states, after stratifying for the type of cancer

Category	Total sample			Expanded states			Non-expanded states
Type of cancer	Pre-Medicaid expansion (2011-2013)	Post-Medicaid expansion (2014-2015)	p value	Pre-Medicaid expansion (2011-2013)	Post-Medicaid expansion (2014-2015)	p value	Pre-Medicaid expansion (2011-2013)	Post-Medicaid expansion (2014-2015)	p value
Bone tumors	(n = 1196)	(n = 869)		(n = 955)	(n = 701)		(n = 241)	(n = 168)
			< 0.001			0.006			0.01
Stage I/II	67% (800)	76% (656)		68% (646)	75% (525)		64% (154)	78% (131)
Stage III/IV	18% (218)	14% (121)		18% (170)	14% (100)		20% (48)	13% (21)
	15% (178)	11% (92)		15% (139)	11% (76)		16% (39)	10% (16)
Unknown stage
Soft-tissue sarcomas	(n = 3880)	(n = 2805)		(n = 3192)	(n = 2270)		(n = 688)	(n = 535)
			0.17			0.31			0.47
Stage I/II	55% (2137)	57% (1608)		55% (1769)	57% (1303)		54% (368)	57% (305)
Stage III/IV	38% (1461)	36% (1010)		37% (1185)	36% (812)		40% (276)	37% (198)
	7% (282)	7% (187)		8% (238)	7% (155)		6% (44)	6% (32)
Unknown stage

Discussion

The results of the current study show that there was a higher proportion of individuals with Medicaid and lower proportion of uninsured patients after the ACA, although the change was more evident after 2014 in states that chose to expand coverage under the Medicaid expansion. Even though Medicaid expansion was responsible for an increased proportion of Medicaid patients, it was not associated with changes in the proportion of uninsured individuals, likely indicating that the slightly lower proportion of uninsured individuals in the post-expansion period may be due to other healthcare reforms that were part of the ACA. Furthermore, we noted an increasing proportion of early-stage diagnoses and decreasing proportion of unknown stage diagnoses for all patients after the ACA was implemented in 2010, although upon stratification by cancer type, stage-shifts were only present for bone sarcomas. Through expansion of coverage and subsidization of private insurance, the ACA successfully decreased the number of uninsured individuals in the United States from 49 million in 2010 to 29 million in 2015 [26]. Despite the widespread coverage gains, the future of the ACA remains unclear because of a lack of complete bipartisan support in Congress. In light of the public debate regarding the possible repeal of the ACA [32], there is a need for a further investigation into the mandate’s effect on access to care for vulnerable patients, including those with cancer. Furthermore, understanding whether the ACA was associated with early-stage diagnoses for bone and soft-tissue sarcomas will be beneficial in discussing the long-term financial ramifications and/or cost-effectiveness of this mandate, particularly given that most sarcomas have a good prognosis if identified and treated earlier.

Limitations

There are several limitations to the study that must be considered when interpreting or translating the findings into clinical practice or health policy. First, the SEER database records insurance status at the time of diagnosis only and does not report changes in insurance status during the follow-up period. A certain subset of these uninsured individuals may have received or opted in for insurance sometime after the diagnosis was made. Vice versa, patients may also become eligible for Medicaid during the course of their treatment. However, given that the purpose of our study was to evaluate insurance status at the time of diagnosis, the use of this variable is justified in the light of the study’s objectives. Similar to other administrative datasets, the SEER database is prone to miscoding, which may lend some bias to the results. However, given a reasonably large sample size of more than 15,000 patients, we do believe that the effects of instances of miscoding bias are somewhat nullified. Because of a limited sample size of pediatric patients and subsequent low power, we were unable to analyze the impact of the ACA on insurance coverage of pediatric patients with bone or soft-tissue sarcomas. Similarly, due to a relatively small sample (n = 100) of uninsured young adults (patients aged 19-26 years) with bone and soft-tissue sarcomas, we were unable to analyze the young adult coverage mandate of the ACA. Given that the objective of the study was to holistically evaluate trends in insurance coverage for bone and soft-tissue sarcomas, we believe the former question may best be suited for a separate study that should be conducted once more data from recent years is available. Data were only available up to 2015, and we were therefore unable to comprehensively evaluate the long-term impact of the ACA on the overall survival of patients with cancer. Because of a limited sample size and a short follow-up period, we also did not evaluate the effect of the ACA on treatment variation (for example, radiotherapy, chemotherapy, and limb-salvage surgery versus amputation). These objectives are beyond the scope of the current study, which aims to assess trends in insurance coverage and stage-shifts, rather than outcomes. Finally, because of a lack of available data, we were unable to ascertain whether individuals enrolling in Medicaid had access to managed-care plans. Lastly, it is difficult to establish causality in this study (that is, it is unknown what healthcare reform that was part of the ACA was directly responsible for the increase in early-stage diagnoses over time). However, such causality is impossible to derive in an observation study, which is further limited by a lack of granular data regarding timing and implementation of different reforms. Not all states chose to expand Medicaid at the same time in 2014, and therefore, considerable overlap in coverage may be present. Unfortunately, we do not have granular data regarding exact dates of reforms, as mentioned previously. Furthermore, the goal of the study was to broadly analyze the role of Medicaid expansion in 2014, rather than granularly differentiate coverages between different states at distinct times. This study is also limited in that we cannot determine whether the changes in the proportion of patients who were insured correlated with a similar improvement in survival. More time and follow-up will be needed to ascertain whether the ultimate goal of improving survival of bone and soft-tissue cancer patients was achieved by ACA

Insurance Coverage Before and After the ACA

The changes in insurance coverage proportions seen in our study can be explained by the various ACA policies. The ACA promotes access to care by allowing individuals to purchase government-regulated plans offered by private payors through official state-level exchanges [11]. It is estimated that roughly 11.5 million individuals are now insured through these exchange plans [2]. Beginning in 2011, employers with fewer than 25 employees with average wages less than USD 50,000 would be eligible for tax credits if they provided health insurance [1]. Another important regulation was prohibiting insurers from denying coverage to individuals with preexisting health conditions (for example, cancer) [10], allowing vulnerable patients to buy plans through exchange marketplaces. By removing the restriction, it is estimated that nearly 100,000 patients with cancer are now enrolled in these exchange plans [16]. Individuals and legal residents are also required to purchase insurance or pay additional taxes. Finally, but most importantly, the ACA offers states the option of expanding Medicaid coverage to individuals and families with an annual income of up to 138% of the federal poverty level [2]. Since 2014, 33 states and the District of Columbia have opted into Medicaid expansion. As evidenced in our study, changes in Medicaid insurance coverage proportions were only noted for states that expanded their Medicaid coverage. These observations are similar to those reported in prior non-sarcoma cancer publications. Using the SEER database, one study [4] analyzed more than 80,000 individuals with newly diagnosed head and neck cancers and observed a 50% reduction in the number of uninsured patients after Medicaid was expanded in 2014. Another study investigated trends in coverage for five common cancers (breast, colorectal, lung, thyroid, and prostate) and noted a 25% to 34% reduction in the number of uninsured patients after Medicaid expansion [30]. In a study examining trends in insurance coverage for colon, lung, and breast cancers, [24] the authors observed that Medicaid expansion resulted in a more than 50% decrease in the proportion of uninsured individuals with colon and lung cancers, but there was no decrease in the proportion of patients with breast cancer. In a separate study evaluating gynecologic malignancies. [23] the authors noted that between 2011 to 2014, the proportions of uninsured individuals with uterine, cervical, and ovarian cancers decreased by nearly 50%. Similar to our findings, these authors observed that the decrease occurred largely among states that adopted Medicaid expansion in 2014. Because a lack of insurance may prevent a patient’s access to timely care and treatment [6, 17] and ultimately impact survival outcomes, the ACA’s Medicaid expansion may be the most useful reform for improving the quality of cancer care delivery. Unfortunately, it is too early to assess whether overall and cancer-specific 5-year survival rates differ between expanded and non-expanded states. Future studies using long-term data will shed more light on whether these healthcare reforms will affect long-term survival rates for bone and soft-tissue sarcomas.

Cancer Stage Shifts Under the ACA

The increasing proportion of early-stage diagnoses and relative decreases in the proportions of unknown and late-stage diagnoses of all sarcomas after the passage of ACA in 2010 is promising. Although we observed a higher proportion of bone sarcomas in the post-expansion (2014-2015) cohort for both expanded and non-expanded states, a similar observation was not seen for soft-tissue sarcomas. The increase in the proportion of early-stage diagnoses of bone sarcomas in both expanded and non-expanded states likely points towards the fact that, apart from Medicaid expansion, other healthcare reforms such as cost-sharing may be responsible for the observed stage-shifts. Staging of bone and soft-tissue tumors often involves multiple tests that can accrue substantial costs. A typical pelvic MRI can cost approximately USD 1000 to USD 5000, depending on the complexity of the scan and hospital location. Similarly, radiography costs approximately USD 200 to USD 350, depending on the location of the scan. The ACA mandates that all participating health plans, including Medicaid, must cover essential health benefits allowing cancer individuals to gain access to cancer testing (for example, radiographic examinations) and initial office visits for diagnostic purposes [3]. Even for individuals with insurance, out-of-pocket payments or copayments may be exceedingly high for specific tests that are necessary for staging tumors. For instance, copays for an MRI can be nearly USD 100, and that of CT images can be nearly USD 200 (https://www.newchoicehealth.com/ct-scan/cost#:∼:text=The%20average%20cost%20of%20a,or%20an%20outpatient%20surgery%20center). To combat the variation in costs of copays and standardize access to care for affected individuals, the ACA introduced cost-sharing subsidization and third-party benefits to offset the burden of out-of-pocket payments [25]. Cost-sharing subsidies were paid to insurance companies to reduce copayments and deductibles for enrollees, thus allowing them to have greater access to services [25]. Although the cost-sharing subsidization regulation largely applies to private exchange plans, Medicaid enrollees in expansion and non-expansion states essentially have no premiums and negligible copayments, allowing them to gain access to essential care. However, most cancer beneficiaries under Medicaid have supplemental managed care plans to ensure that costs and care for certain expensive diagnostic tests are covered [11]. All of these reforms could ensure that affected individuals have access to care when becoming symptomatic, thus yielding a greater proportion of early-stage diagnoses for bone sarcomas between 2014 and 2015 compared with 2010 and 2013.

Conclusion

In conclusion, as we anticipated, we observed that access to cancer care for patients with primary bone or soft-tissue sarcomas improved after the ACA was introduced, as evidenced by a decrease in the proportion of uninsured patients and corresponding increase in Medicaid coverage. These outcomes were only demonstrated in states that adopted the 2014 Medicaid expansion. The ACA may also yield earlier diagnoses by improving access to care, as evidenced by an increase in the number of early-stage diagnoses and a decrease in the number of unknown-stage diagnoses after 2010 for patients with bone but not soft-tissue sarcomas. The findings of this study support further discussion on the future of the ACA mandate, particularly in the current political climate, to ensure cancer care is equitable and accessible to all patients.