Abstract
Medicare site-dependent payments and the 340B Drug Pricing Program both incentivize vertical integration between hospitals and physician practices. Medicare pays higher rates to hospital outpatient departments (HOPDs) than to physician offices, incentivizing hospitals to acquire and convert standalone physician practices into off-campus HOPDs. 340B hospitals earn enhanced profits from outpatient drug delivery, which may further incentivize acquisition of new sites. We used Medicare fee-for-service claims to describe trends between 2016 and 2022 in services provided at off-campus HOPDs, an indicator of vertical integration, stratified by hospital 340B participation status. 340B hospitals delivered a higher and growing percentage of outpatient services at off-campus HOPDs than non-participating hospitals. Drugs accounted for a larger percentage of off-campus HOPD claims at 340B hospitals. At off-campus HOPDs paid site-neutral rates, service volume increased faster among 340B hospitals. Despite recent reforms, 340B hospitals may be more incentivized to pursue vertical integration than other hospitals.
Keywords
Introduction
Medicare payment rules provide incentives for vertical integration between hospitals and physician practices (Medicare Payment Advisory Commission [MedPAC], 2023). Under the Outpatient Prospective Payment System (OPPS), Medicare typically reimburses hospital outpatient departments (HOPDs) at a higher rate than physician offices for the same services (Bulat & Brake, 2024; MedPAC, 2023). Hospitals are incentivized to acquire standalone physician practices and convert them into off-campus HOPDs that obtain these higher reimbursement rates, leading to higher health care costs for Medicare, commercial payers, and patients (Bulat & Brake, 2024; Dranove & Ody, 2019; Maughan et al., 2025; Neprash et al., 2015; Post et al., 2021).
Discounted outpatient drug acquisition costs under the 340B Drug Pricing Program (“340B”) enhance the profitability of administering drugs in outpatient settings for participating non-profit and public hospitals, which may create an additional incentive for vertical integration with physician practices (Health Resources & Services Administration [HRSA], 2025; Keene et al., 2023). 340B hospitals are estimated to receive discounts of 29% off drug average sales prices (ASP) (Neuman & Ray, 2024), yet insurers and patients may still be billed for the full price (Bond et al., 2023; Desai & McWilliams, 2018). Medicare reimbursement for these outpatient drugs is equal to the ASP plus 6% regardless of 340B participation status (Robinson et al., 2024). With one exception (Li & Xu, 2022), evidence suggests 340B participation is associated with utilization of higher-cost drugs and more drugs (Bond et al., 2023; Thomas & Schulman, 2020; United States Government Accountability Office [GAO], 2015).
Vertical integration may be particularly attractive to 340B hospitals due to the incentives of both site-dependent payments and outpatient drug profitability. 340B hospitals have acquired off-campus sites in wealthier neighborhoods (Conti et al., 2019; Liu et al., 2025), and may contribute to shifts in the site of care from physician offices to HOPDs (Jung et al., 2018). Medicare outpatient drug spending is increasingly shifting toward the HOPD setting and away from non-hospital outpatient settings (Nguyen et al., 2023). Existing research raises questions about whether hospitals receiving 340B drug discounts in addition to site-dependent payments engage in more vertical integration with physician practices compared to hospitals not participating in 340B.
Reforms implemented to improve these payment incentives have been incremental, but policy interest persists (Table 1, timeline in Supplemental eFigure 1). Payments to off-campus HOPDs newly established or relocated after November 2015 were made “site neutral” under the Bipartisan Budget Act (BBA) of 2015, meaning payment rates to these HOPDs were reduced to the Medicare physician fee schedule rates (Bulat & Brake, 2024; Centers for Medicare & Medicaid Services [CMS], 2016; MedPAC, 2023). Due to exceptions for on-campus and pre-existing off-campus HOPDs, the policy affected less than 2% of HOPD spending (Bulat & Brake, 2024; Post et al., 2025). However, CMS expanded the policy in 2019 by applying site-neutral rates to clinic visits provided at pre-existing off-campus HOPDs, and proposed a new rule that would do the same for drug administration services beginning 2026 (CMS, 2018, 2025a, 2025b; Lower Health Costs for Seniors Framework, 2024).
Applicability of Site Neutral and 340B Payment Policy Changes to Off-Campus HOPDs.
Green indicates the site-neutral policy and blue indicates the 340B payment policy, while gray indicates the payment policy change did not apply. bMost outpatient drugs are reimbursed by Medicare at the average sales price (ASP) + 6%. Starting in 2018, Medicare reimbursement for 340B drugs was decreased to ASP - 22.5%. This change lasted until the court ruling in AHA v. Becerra prompted CMS to revert to ASP + 6% for 340B drugs, effective Sept 27, 2022.
In 2018, CMS reduced drug payments to 340B hospitals from ASP plus 6% to ASP minus 22.5% to better align payment with acquisition cost, but this was reversed in September of 2022 by the Supreme Court (CMS, 2023). The Trump Administration expressed renewed interest in lowering 340B Medicare drug payments in an April 2025 executive order, but details have yet to be announced (Trump, 2025). Senator Bill Cassidy, Chair of the U.S. Senate Health, Education, Labor, and Pensions Committee, has also advocated for 340B reforms to increase transparency and direct benefit to patients (U.S. Senate Committee on Health, Education, Labor & Pensions [HELP], 2025). The pharmaceutical industry organization PhRMA has also revamped its campaign to persuade Congress to curb 340B markups (Muoio, 2025). However, hospitals consistently oppose 340B reform, and the hospital financial challenges likely to result from the One Big Beautiful Bill’s cuts to Medicaid funding may strengthen the case against reform (American Hospital Association [AHA], 2025; Rosen, 2025).
Only one other study has examined the dual incentives for hospital-physician vertical integration provided by site-dependent payments and 340B drug profitability; it did so in the context of cancer care and before recent payment policy changes (Jung et al., 2018). Understanding trends in hospital behavior amid recent site-neutral and 340B policy changes can inform both future research and additional federal site-neutral payment and 340B reforms. We quantified trends in Medicare service volume at off-campus HOPDs, an indicator of vertical integration between hospitals and physician practices, by hospital 340B participation status. Off-campus HOPD service volume is an indicator of vertical integration because these sites are often former physician offices that were acquired by hospitals and converted into HOPDs (Capps et al., 2018). Site-neutral reforms introduced claims modifiers delineating between pre-existing and newly established off-campus HOPDs (as of November 2015); therefore, we could separately observe service delivery changes within existing sites and growth in relatively new sites that reflect likely recent vertical integration with freestanding clinics or offices. We hypothesized that 340B hospitals would have higher overall off-campus utilization driven by outpatient drug administration. We also expected 340B hospitals to have faster-increasing off-campus HOPD claims volume at site-neutral locations compared to hospitals not participating in 340B, suggesting that site-neutral payment reforms may be a less effective deterrent of vertical integration among 340B hospitals.
New Contribution
Prior literature has largely examined the 340B Program and Medicare site-dependent payments separately. To our knowledge, this is the first study using Medicare claims modifiers to examine an indicator of hospital vertical integration behavior stratifying by exposure to both 340B and site-dependent payment incentives.
Method
Data
This study used four data sources: (a) a 20% random sample of Medicare fee-for-service claims to calculate service volume in settings classified as hospital outpatient departments, (b) the RAND version of the Medicare Hospital Cost Report Data to obtain hospital characteristics (RAND, 2025), (c) the Health Resources and Services Administration (HRSA) 340B Office of Pharmacy Affairs Information System (340B OPAIS) to identify hospitals participating in 340B (HRSA, n.d.), and (d) the CMS OPPS Quarterly Addenda files to obtain Ambulatory Payment Classification (APC) codes (CMS, 2025c).
Sample
The sample included short-term hospital outpatient claims from January 1, 2016, to December 31, 2022 (Supplemental eFigure 2). Claims were identified by a unique claim identification number. We included claims with payment amount greater than $0, with an assigned APC code, and that were paid under the OPPS in all states and the District of Columbia, except Maryland. Non-OPPS claims, including claims from Maryland hospitals and other OPPS-exempt hospitals, were excluded because the reimbursement models for these hospitals may provide different incentives for off-campus service delivery.
We divided the hospitals into three groups to distinguish between: (a) 340B hospitals that initiated 340B participation prior to 2016 (“active 340B”), (b) non-profit or government hospitals that were potentially eligible for 340B but never participated as of December 31, 2022 (“non-340B”), and (c) 340-ineligible for-profit hospitals (“for-profit”). More information on 340B eligibility can be found in eMethods 1. To avoid cases where hospitals switched between groups during the study period, we excluded hospitals that initiated 340B participation during the study period (N = 297) and any 340B hospitals that terminated their 340B participation between their initial participation date and the end of the study period (N = 327).
Outcomes
Hospital outpatient services can occur in an outpatient department on the hospital’s main campus or at off-site outpatient departments. We aimed to measure the extent to which hospitals deliver services at HOPDs not located on the main hospital campus (“off-campus HOPDs”). Off-campus HOPDs are often former freestanding physician offices that were acquired by a hospital (Capps et al., 2018), therefore, service delivery at these sites serves as an indicator of vertical integration between hospitals and physician practices. The primary outcome was the aggregate proportion of outpatient claims in each hospital 340B participation group occurring at off-campus HOPDs per quarter. We also examined trends in off-campus drug-related claims and clinic visit claims.
Off-campus claims were identified using claims billing code modifiers. The “PO” modifier became mandatory for all off-campus HOPDs according to CMS payment rules on January 1, 2016 (CMS, 2014). In 2017, the first year of BBA site-neutral payments, CMS required facilities exempt from site-neutral payment to use the “PO” modifier and facilities subject to site-neutral payment to use “PN” (CMS, 2016). “PO” therefore represents service volume at pre-existing off-campus HOPDs and “PN” represents service volume at newly acquired, developed, or relocated off-campus HOPD sites after November 2015 (CMS, 2016; McGraw Walsh et al., 2017). Conversely, on-campus claims were those without modifiers “PO,” “PN,” and “ER” (which was introduced in 2019 to identify emergency services at off-campus HOPDs). This approach to identifying off-campus claims and distinguishing between sites subject to and exempt from site-neutral reforms has been used in prior literature (Post et al., 2025). Starting in 2019, CMS extended site-neutral payment to clinic visits at exempt off-campus HOPDs, but these facilities were still instructed to continue billing clinic visit claims using “PO” (CMS, 2018). Clinic visit claims were identified using APC code 5012 (CMS, 2025c).
We also examined drug and drug administration claims (collectively “drug claims”). Claims for outpatient drugs were identified as those with an APC code associated with the status indicator for pass-through drugs (G) or non-pass-through separately payable drugs (K) (CMS, 2025c). These APCs represent higher-cost drugs that are paid at ASP plus 6% (“separately payable drugs”) and for which the administration fee is billed separately (MedPAC, 2021). Among 340B hospitals, this group of drugs was affected by the 2018 340B drug payment cuts, which reduced separately payable drug reimbursements to ASP minus 22.5% (CMS, 2023; MedPAC, 2021). Drug administration claims were those with APC code 5691, 5692, 5693, or 5694, which correspond to drug administration Levels 1 through 4. These APCs represent either an OPPS bundled payment for both the drug and administration fee, or the administration fee associated with a separately payable drug. This group was affected by the 2017 site-neutral reforms, which aligned payment rates to the physician fee schedule, regardless of 340B participation status (CMS, 2016). All other claims with an APC code not meeting our definition for drug claims were classified as non-drug claims.
Analysis
We conducted a descriptive analysis to examine trends in off-campus utilization, overall and by facility site-neutral status, comparing 340B hospitals with non-340B and for-profit hospitals. We reported the aggregate proportion of off-campus claims among all hospital outpatient claims by the 340B group per quarter. We chose this measure over reporting means or medians per hospital because the number of off-campus claims per hospital was right-skewed, with many hospitals having zero off-campus claims in a given quarter. Aggregate group-level proportions avoid these distributional challenges and better reflect the market-level allocation of outpatient services to off-campus sites among the three groups of hospitals. We also conducted a sensitivity analysis stratifying the hospitals by quartile of number of beds to account for differences in hospital size between the groups (eMethods 2). Analyses were conducted in R Version 4.4.0 (R Foundation for Statistical Computing). This study followed STROBE cohort reporting guidelines (von Elm et al., 2007).
Results
Hospital Characteristics
The sample included 2,550 hospitals with 160,760,016 outpatient claims. Among the hospitals, 930 participated in 340B, 844 were potentially eligible non-profit or government hospitals that never participated in 340B, and 776 were 340B-ineligible for-profit hospitals (Table 2). 340B hospitals were more commonly private non-profits and teaching hospitals compared to the other groups. 340B hospitals also had more beds, larger Medicare claims volume, and higher disproportionate share (DSH) percentages than the other groups. Stratifying by quartiles of hospital bed size produced groups with a more similar number of beds, baseline revenue, and baseline claims volume (Supplemental eTables 2.1–2.4).
Baseline Hospital Characteristics (2016).
Note. DSH: disproportionate share hospital (DSH ≥11.75% is the threshold for 340B eligibility for most hospitals). Active 340B: hospitals initiating 340B participation prior to January 1, 2016, and remaining active during the study period. Non-340B: non-profit and government hospitals that are potentially eligible for 340B but have never participated. For-profit: for-profit hospitals, which are ineligible for 340B. Claims counts refer to outpatient claims.
Source: 20% random sample of Medicare FFS claims.
Off-Campus Service Delivery by Site-Neutral Status
Among the three groups of hospitals, 340B hospitals had the highest percentage of outpatient services occurring at off-campus HOPDs across all years (Figure 1). This was true when aggregating all off-campus HOPDs and when stratifying by facility site-neutral status. When stratifying by quartile of hospital bed size, this remained true for hospitals above the median size. In the first quartile, for-profit hospitals had the highest percentage of outpatient services occurring off-campus, and in the second quartile, 340B and non-340B had similar levels and were both higher than the for-profit group (Supplemental eFigure 3). Overall, 340B hospitals increased the percentage of outpatient services delivered in off-campus HOPDs from 18.2% in the first quarter of 2016 to 24.1% in the last quarter of 2022. In contrast, this figure increased much less at non-340B hospitals (16.6–18.3%) and declined slightly at for-profit hospitals (8.0–7.5%). The percentage of services delivered off-campus at HOPDs exempt from site-neutral payment remained flat at 340B hospitals and non-340B hospitals and slightly declined at for-profit hospitals. 340B hospitals had the fastest increase in the percentage of outpatient services delivered in new site-neutral off-campus HOPDs compared to the other hospitals, increasing from zero in 2016 (before the site-neutral policy took effect) to 5.8% by the end of the period (compared to 2.2% and 0.6% among non-340B and for-profit hospitals, respectively).
Aggregate proportion of outpatient claims in off-campus HOPDs by site-neutral and 340B participation status per quarter.
Contribution of Drug Claims
When examining the composition of claims across all off-campus HOPDs, the proportion of off-campus claims attributed to outpatient drugs and drug administration (collectively “drug claims”) versus non-drug services and items differed by 340B participation status (Figure 2). Among 340B hospitals, drug claims accounted for 24% of off-campus claims, compared to 13% and 7% among non-340B and for-profit hospitals, respectively. Drug claims were a growing share of off-campus claims over the study period among 340B hospitals (4% average annual increase), whereas the drug share of off-campus claims increased more slowly (1% average annual increase) among non-340B hospitals and declined by 2% annually, on average, among for-profit hospitals. The stratified analysis by hospital size showed similar results (Supplemental eFigure 4).
Contribution of drug claims to off-campus claims volume, 2016 to 2022.
Off-Campus Claims by Service Type
Trends in outpatient drug delivery were of interest due to the financial incentives of the 340B program. Clinic visits were also important to examine, given that they became newly subject to site-neutral payment at exempt off-campus HOPDs in 2019. As a share of outpatient services, clinic visits at off-campus HOPDs exempt from site-neutral payment (which became subject to site-neutral payment in 2019) declined among 340B (9.3–7.6%) and non-340B (7.8–5.9%) hospitals between 2016 and 2022 (Figure 3). Clinic visits increased slightly in 2021 during the COVID-19 pandemic, but resumed declining in 2022. Among 340B hospitals, drug claims as a share of outpatient services at exempt off-campus HOPDs increased slightly (3.4% in the first quarter of 2016 to 4.9% in the last quarter of 2022). Among non-340B hospitals, this figure remained flat at approximately 2%. At site-neutral off-campus HOPDs, clinic visits and drug claims increased faster among 340B hospitals compared to non-340B hospitals. By the last quarter of 2022, the percentage of outpatient services attributed to clinic visits at site-neutral HOPDs was 2.9% among 340B hospitals versus 0.9% among non-340B hospitals, and drug claims from these HOPDs represented 1.7% (340B) and 0.2% (non-340B) of outpatient claims. These trends were largely consistent across hospital bed size strata, notably indicating faster increases in site-neutral HOPDs among 340B compared to non-340B hospitals (Supplemental eFigure 5). For-profit hospitals had a substantially lower percentage of both clinic visits and drugs delivered in off-campus HOPDs compared to other hospitals and more stable time trends (Supplemental eFigure 3).
Aggregate proportion of off-campus claims by service type per quarter.
Discussion
In this descriptive analysis of trends in Medicare outpatient claims from 2016 to 2022, we found that 340B hospitals delivered a higher and more rapidly increasing percentage of outpatient services at off-campus HOPD sites than non-340B and for-profit hospitals. Drugs constituted a larger and faster-growing share of off-campus services at 340B hospitals than at other hospitals. At newly established or relocated off-campus HOPDs subject to site-neutral payment, 340B hospitals had faster increases in the relative service volume compared to non-340B hospitals and for-profit hospitals. This suggests that 340B hospitals have different incentives for vertical integration compared to hospitals not participating in or ineligible for 340B, and may respond differently to site-neutral payment reforms. 340B hospitals are a particularly important group to target for reforms curbing vertical integration, since these hospitals accounted for more than half of the volume of off-campus service delivery observed in this study. Considering the influence of multiple intersecting payment policies is important for policymakers developing new Medicare payment reforms.
By the end of the study period, 24% of 340B hospital outpatient services were delivered at off-campus HOPDs, compared to 18% and 8% among non-340B and for-profit hospitals. This difference was driven by larger hospitals. 340B hospitals increased the share of outpatient services delivered off-campus by 6 percentage points over the period, whereas non-340B and for-profit hospitals had minimal change. This finding adds to the literature, indicating that 340B hospitals have tended to add new sites and integrate with physician practices (Conti et al., 2019; Desai & McWilliams, 2018). However, overall off-campus volume trends could reflect both changes in delivery within existing sites that may have been acquired in the past or the acquisition of new sites; therefore, it was important to also examine volume within newly established sites.
Claims from new off-campus HOPDs subject to site-neutral payment increased over time in all hospital groups starting in 2017, and fastest among 340B hospitals regardless of hospital size. The faster rate of increase in claims subject to site-neutral payment among 340B hospitals reflects how site-neutral policies may be less of a deterrent for vertical integration among 340B hospitals than among non-340B hospitals due to the additional incentive for increased outpatient drug administration. Further, the faster increase in claims subject to site-neutral payment among 340B hospitals was not unique to drugs: 340B hospitals also increased non-drug service delivery at a faster rate in site-neutral off-campus HOPDs compared to non-340B hospitals. This is consistent with prior findings suggesting that 340B hospitals increase delivery of outpatient drugs as well as other related medical services (Jung et al., 2018).
Despite the increases in site-neutral service volume, the number of claims subject to site-neutral payment remains low, and the reforms have not slowed spending and vertical integration (Post et al., 2025). This may be because CMS rules leave a loophole allowing hospitals to purchase physician practices and employ those physicians at existing off-campus HOPDs that are exempt from site-neutral payment (CMS, 2016; MedPAC, 2023). Employment of physicians, rather than acquisition of new facilities, is an increasingly common approach to vertical integration (Post et al., 2022). This underscores the importance of a new proposal by Senators Bill Cassidy and Maggie Hassan to extend site-neutral payment to all off-campus HOPDs to reduce the incentives for increasing the number of physicians and services at exempt off-campus HOPDs (Lower Health Costs for Seniors Framework, 2024; MedPAC, 2023).
Drug claims contributed to an increase in 340B off-campus HOPD claims volume, consistent with prior literature indicating that 340B participation incentivizes higher outpatient drug utilization (Bond et al., 2023; Desai & McWilliams, 2018). Outpatient drugs were a higher and increasing share of off-campus services at 340B hospitals compared to non-340B and for-profit hospitals across hospital bed size strata. This may reflect 340B hospital prioritization of profitable service lines like oncology drug administration over less profitable services like psychiatry (Owsley et al., 2024). This growth in off-campus delivery of outpatient drugs in 340B hospitals occurred despite the reductions in 340B drug payments occurring between 2018 and September 2022. Policymakers have recently proposed reinstating the reduction in 340B drug payments to better align with acquisition costs (Trump, 2025). While reducing 340B drug reimbursement would lower Medicare drug spending, our findings indicated there was still higher relative growth in drug and other service delivery at newly established off-campus HOPDs of 340B hospitals during the period of lower drug reimbursement (2018–2022). This suggests that reinstating 340B drug payment cuts may not curb growing utilization and vertical integration among 340B hospitals. Expansion to new sites may even be used as a strategy for accelerating volume increases in response to payment shocks. Alternatively, hospitals may have expected the payment policy would be reversed by the courts, which was the ultimate outcome (CMS, 2023).
Clinic visits declined at exempt off-campus HOPDs, possibly due to the 2019 expansion of site-neutral payment to these services (CMS, 2018). The declines preceded 2019, suggesting hospitals may have enacted anticipatory changes when initial site-neutral reforms were implemented in 2017. CMS intended to reduce excessive spending and shift clinic visits, which accounted for one-third of outpatient services, toward less costly physician offices. Clinic visit payments decreased from $116 to $46 (and copayments from $23 to $9), which may have prompted hospitals to offer fewer of these visits (CMS, 2018). In this environment, hospitals may favor shifting service mix toward complex services and drugs to insulate against future site-neutral policies targeting services that can be delivered in both HOPD and office settings (Moody & Enders, 2025). Consistent with CMS goals and Medicare Payment Advisory Commission recommendations (CMS, 2018; MedPAC, 2023), targeting site-neutral payment toward services that can be delivered safely in both HOPDs and physician offices may be an effective way of reducing unnecessary spending. However, additional reforms of this nature would occur amid the effects of the July 2025 One Big Beautiful Bill Act on hospital finances. Due to the increase in uninsured patients resulting from this legislation, hospitals are expected to face lost revenue (Rosen, 2025; Strul et al., 2025), reducing the political feasibility of additional site-neutral reforms that would further reduce hospital revenue.
This study has several limitations. First, the analysis is descriptive and does not measure causal relationships. Second, there may have been inconsistencies in the use of the claims modifiers as hospitals adjusted to the new requirement; however, there was no evidence of noncompliance found in the literature. Third, the use of off-campus claims volume as an indicator of vertical integration is incomplete, as we could not completely isolate changes in the number of sites from changes in volume at existing sites, such as through physician employment changes. Fourth, while this study examines Medicare claims, payer mix is relevant to hospital decision-making and negotiated rates with commercial payers may also influence a hospital’s propensity for vertical integration behavior.
Conclusion
Between 2016 and 2022, 340B hospitals delivered a higher percentage of outpatient services in off-campus HOPDs compared to non-340B and for-profit hospitals. Off-campus service delivery increased fastest among 340B hospitals, driven by faster growth in newly established site-neutral off-campus HOPDs. Despite site-neutral reforms and temporarily lower 340B drug payments, 340B hospitals may still be more incentivized to pursue vertical integration with physician practices compared to other hospitals.
Supplemental Material
sj-docx-1-mcr-10.1177_10775587261453476 – Supplemental material for Changes in Service Delivery Following Site-Neutral Payment Reforms by Hospital 340B Status
Supplemental material, sj-docx-1-mcr-10.1177_10775587261453476 for Changes in Service Delivery Following Site-Neutral Payment Reforms by Hospital 340B Status by Ilina C. Odouard, Julia C. Thome and Melinda B. Buntin in Medical Care Research and Review
Footnotes
Acknowledgements
The authors are grateful for assistance from Lauren Da Fonte, Pikki Lai, and Kathryn Linehan.
Ethical Approval
This study received ethical approval from the Johns Hopkins University IRB (approval # IRB00026546: Evaluating Medicare Policy Options) on November 4, 2023.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grant number 25092 from the Peterson Center on Healthcare. Ms. I.C.O. was supported by grant number T32HS000029 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. M.B.B. is an unpaid member of the Board of Harvard Medical Faculty Physicians, on the Advisory Board of the Peterson Health Technology Initiative, and Senior Associate Editor for JAMA Health Forum.
Data Availability Statement
The data are not permitted to be shared publicly according to the Centers for Medicare and Medicaid Services rules.
Supplemental Material
Supplemental material for this article is available online.
References
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.
