Crunching the Numbers: Data and Platforming Value in Medicaid

INTRODUCTIONIn 2015, the Ohio Department of Medicaid processed more than 3 million claims through its online portal, the Medicaid Information Technology System (MITS), representing nearly $960 million in reimbursement and providers submitted more than 24 million eligibility inquiries.[1] Additionally, with nearly $48,000 a minute reimbursed to Ohio’s nearly 90,000 providers, the Ohio Medicaid program’s integrity is paramount, with the federal government, the Ohio Attorney General, the Auditor, the Department of Medicaid and local County Departments of Job and Family Services. These agencies are working to ensure payments are made legally and in accordance with law. Between claims, eligibility processing, auditing and the provision of care, millions of pieces of data are flowing back and forth between patients, governments and providers.

Data in public policy is simultaneously ambiguous and ubiquitous, cited as a critical need, an asset worth protecting and an elusive salve to the stereotypical inefficiencies of bureaucracy.

Data in public policy is simultaneously ambiguous and ubiquitous, cited as a critical need, an asset worth protecting and an elusive salve to the stereotypical inefficiencies of bureaucracy. However, in an age defined by the ability to collect and disseminate data, how has Ohio’s largest single source of coverage dealt with data? In what ways has data refashioned Medicaid, and what are the major trends informing these changes? The following Issue Brief seeks to answer some of these questions and attempts to bring clarity to the intersection of Medicaid and data. To do this, the paper will look at four major domains of data: eligibility, claims, auditing and electronic health records. While this is by no means comprehensive, the last few years of investment by Governor John Kasich's administration in these areas of data infrastructure can be seen as the foundation upon which value-based reform will be built.KEY TAKEAWAYS

• Systems of eligibility, claims and health records have grown significantly and proliferated the availability of data in Medicaid
• Data in Medicaid is currently leveraged to bolster efforts to reduce fraud and waste between multiple levels of government
• Value-based reform can be seen as a data project, and the government plays a key role in platforming the use of data in creating value-based reimbursement policy

[bctt tweet="How does #Ohio #Medicaid deal with data? What impact can it have? Learn more here " username="CommunitySols"]ELIGIBILITYThe Governor’s Office of Health Transformation (OHT) has not minced words when it comes to the eligibility process for Medicaid. As is a hallmark of OHT, the office broadcast its intentions since its inception with the following mission statement:“Eligibility processes for health and human services programs in Ohio are fragmented, overly complex, and rely on outdated technology. In 2013, the Governor’s Office of Health Transformation initiated an eligibility modernization project to simplify client eligibility based on income, streamline state and local responsibility for eligibility determination, and modernize eligibility systems technology. The goal is to improve the consumer experience and significantly reduce the costs associated with these processes.” [2]On a fundamental level, the Department of Medicaid is obligated through its agreement with the federal government to ensure that individuals eligible for Medicaid are determined as such; this responsibility is the core intent of an entitlement. The Centers for Medicare and Medicaid Services (CMS) relies on states to regularly report on eligibility determination and to meet standards of performance in regards to certification.[3] The executive branch is responsible for ensuring that the benefit is appropriately administered and, as such, it regularly reviews the ways in which eligibility is applied as a matter of regulatory oversight.Historically, eligibility determination was conducted through an in-person process, with County Departments of Job and Family Services (CDJFS) acting as the entity responsible for processing. While there was a legacy eligibility system in the Client Registry Information System Enhanced (CRIS-E), this system was universally seen as complex, outdated and inefficient, but the cost of replacing it was also seen as too high. To improve the system, the governor’s first budget called for the simplification of the eligibility system and accessed federal dollars made available through the Affordable Care Act (ACA) to finance the majority of the work. The goal of building a new system was both to create a product wherein the process of applications could be decentralized at the level of submitting applications (individuals submitting an application through a portal) and to automate the eligibility process and, ostensibly, to better communicate between multiple social service programs (cash assistance, food stamps).The new system of eligibility is called the Ohio Integrated Eligibility System (OIES). OIES is set to come online, statewide, in 2018. The Center for Community Solutions has written about its nascent development and will continue to do so as it is rolled out.[4] Critically, the role of in-person case management through implementation would be theoretically diminished. In business terms, vertical integration is a process of merging two parts of a specific process that are at different stages in production. With greater data overlap and less in-person need, the state is essentially vertically integrating the traditional eligibility functions of county government and automating it, thus, potentially saving expense. In other words, the state is not only covering individuals, it is also assuming the role of determinations by automating them.

In other words, the state is not only covering individuals, it is also assuming the role of determinations by automating them.

However, the functionality of this system remains to be seen, especially in regards to recent efforts to enact work requirements for the Medicaid expansion population. Ohio is pursuing an 1115 Demonstration waiver that would require adults enrolled in the Medicaid expansion to meet a standard of “community engagement,” which primarily includes employment. We have written extensively about this waiver, but it should be known that the legality of such a waiver is in question, nationally.[5] While there are a number of qualifying factors which would exempt individuals from the work requirement, the individual efforts enrollees have to go through is hard to predict. According to the state’s application to CMS, the impact on individuals of the new requirements would be slight and OIES would automate as much of the eligibility process as possible. However, with the expectation that individuals self-report any changes in financial circumstances, it remains to be seen how an automated system responds to real-world beneficiary need.[6] What’s more, automating eligibility processes can lead to disastrous results for beneficiaries. In Indiana, nearly a million beneficiaries were wrongly kicked out of the Medicaid program as the result of an automated system, an example made famous by Virginia Eubanks’ “Automating Inequality.”[7] While this is just one example, it demonstrates the need for states to be conscious of the ways in which data can be used discriminatorily.

Automating eligibility processes can lead to disastrous results for beneficiaries.

CLAIMS

Medicaid is a publicly subsidized insurance product. An essential function of the Medicaid programs, then, is to provide medical assistance through the reimbursement of providers. To do this, the state and federal government share in the expense associated with any given population or effort. This process is known as the federal medical assistance percentage (FMAP) and is built on the average per capita income of any given state, with wealthier states receiving less matching funds than those states with lower levels.[8] In 2011, OHT continued the work of the Taft and Strickland administrations in implementing a new claims payment system to handle the nearly 64 million claims processed for Ohio’s 90,000 providers to care for the state's 3 million recipients. [9]Ohio has implemented a new system of claims processing (adjudication) called the Medicaid Information Technology System (MITS). Through MITS, providers can be enrolled, requests for prior authorization can be made, claims can be submitted and any given provider can interact with the state asynchronously.[10] As with the transition of the eligibility system, MITS can be seen as a streamlining of an organizational function of the Department of Medicaid. Where there was once a largely paper-based system, there is now an adjudication function of Medicaid that can process things more quickly and without the need of a person to intervene in each case. What’s more, this consolidation of claims information means that the state can now review and identify more readily as the submission and function of payment is more uniform.It should be noted, however, that the majority of payments no longer flow through the Department as a fee for service (FFS) claim. Instead, most payments are made indirectly through the State’s Managed Care Organizations (MCOs). MCOs are privately contracted insurance companies which manage the benefits and payments to providers on behalf of the state. MCOs are subject to all the same standards of payment as the state, though they are distinct in how they are able to pay in a few key ways. First, MCOs have utilization controls, such as prior authorization, which allow them to withhold payment for a service until it is deemed appropriate. Second, MCOs are not obligated to pay the same rate as FFS, meaning the rate is the matter of a negotiated contract between any provider and a given MCO. Third, MCOs have more flexibility in paying for things that are not traditionally covered by Medicaid, though they still have to conform to federal laws. Last, the amount of money paid to the MCOs by the state must be actuarially sound, meaning an independent actuary certifies the amounts are sufficient to ensure access for beneficiaries.

Given the large investment Medicaid represents to both the state and federal governments, there is a tremendous focus on reducing waste, fraud and abuse.

The state does have access to the claims being paid by MCOs through “encounter data.” Additionally, the state is able to measure each plan’s performance in achieving quality through claims review and medical measurement, though that will be explored in a subsequent blog in more detail. Interestingly, while MCOs have tremendous flexibility in designing independent claims systems, they typically pay for services as a percentage of FFS. Depending on how the MCO is responding to its contract with the state and maximizing its revenue capture, it may change the ways in which it contracts with providers. In other words, if the MCOs have identified ways to make or earn more money, they can adjust their contracts with medical professionals to maximize that opportunity.AUDITINGGiven the large investment Medicaid represents to both the state and federal governments, there is a tremendous focus on reducing waste, fraud and abuse. This process of ensuring that payments are made as intended is often referred to as Program Integrity which, as CMS defines it, can encompass auditing, identification of overpayments, education and recovery. [11]There are a number of federal guidelines that outline the integrity efforts of the government, many of which are executed or overseen by the Ohio Department of Medicaid. In 2014, ODM created the Bureau of Program Integrity, which sought to coordinate efforts for integrity both within Ohio’s program and between external partners like the federal government. According to a 2015 report provided to the Joint Medicaid Oversight Committee, these activities include provider enrollment and support, pre-payment review, post-payment review, contract management, eligibility testing, sub-recipient monitoring and training.[12] Recently, the U.S. Government Accountability Office published a list of recommendations on the ways that CMS can improve its oversight of states, stressing the need for increased collaboration between levels of government and increasing the participation of states in these activities.[13]Beyond ODM, there are roles for other statewide offices and local governments in the integrity process. The Auditor of State ensures there is compliance with federal law, typically through the retroactive review of expenditures. In addition to Medicaid this offers another layer of review to Ohio’s six current program integrity groups and relies on data from ODM resources such as MITS. [14] In these reviews, the Auditor looks for irregularities in the data, comparing recipient and provider information. Some of the most typical kinds of fraud include billing for services that were never rendered, “upcoding” (which is the practice of billing for a higher-priced treatment than what was actually provided) or providing unnecessary services. In all cases, this fraud is conducted on the provider level, not by the consumer directly.In addition to the Auditor, the Attorney General (AG) and the local Department of Job and Family Services also work on integrity. For the AG, this often involves fraud detection through its Medicaid Fraud Control Unit and prosecutes civil and criminal offenses, including cases of patient neglect or abuse, the work of which is established in a contract with ODM. According to a 2017 report from the AG, this included more than 1,000 investigations.[15] Cases of consumer-based fraud are usually referred to local JFS Departments. Sometimes this can involve the coordination of fraud between a consumer and a provider, but mostly JFS is responsible for ensuring individuals are accurately reporting information relative to eligibility. Beyond all of this, individuals, providers and patients are encouraged to report fraud to the state in the event they are witness to any fraud, waste or abuse taking place.The ACA increased the volume of Medicaid payments through increasing the population of recipients via the optional expansion of benefits to non-disabled adults and a number of integrity provisions such as enhanced provider screening and expanded audit efforts. In addition to the ACA, the 2009 Health Information Technology for Economic and Clinical Health (HITECH) Act also created a number of incentives to leverage the use of data through the incentivizing of electronic health records (EHR). A recent Kaiser Family Foundation review of all 50 states’ Medicaid programs shows many of the integrity activities of states have leveraged this increased availability of data and funding to create advanced analytics as a component of program integrity. This has included predictive modeling, forensic reviews of procurement, streamlining provider enrollment and the exchange of data with other auditing bodies such as the Federal Bureau of Investigation and the Office of the Inspector General (OIG). [16] With these investments, both the federal government and the states have evolved their efforts in program integrity, using the data infrastructure financed by the ACA and HITECH to support targeted integrity interventions.ELECTRONIC HEALTH RECORDSElectronic Health Records (EHR) have not always been a part of the health care delivery system. In fact, fewer than 25 percent of providers had EHR in 2004, a number which had more than tripled by 2015 with an adoption rate of nearly 87 percent.[17] While the benefits of EHR may be intuitively apparent to providers, the HITECH Act did a lot to encourage their adoption and use.In 2009, as a part of the American Recovery and Reinvestment Act (ARRA/the “stimulus”), HITECH helped finance the purchase and incent the use of EHR. To receive benefits, providers had to show that they were “meaningfully using” the records, which included activities such as e-prescribing, electronic exchange and the submission of quality measures.[18]  According to the Congressional Budget Office, these payments are roughly $30 billion, with about $12.4 billion directed through Medicaid between 2011 and 2019. These funds are 100 percent federally sourced. In a 2012 OIG report, Ohio made nearly $193 million in these payments during calendar years 2011 and 2012, with a net overpayment of $524,162 (less than .002 percent). Not all Medicaid providers are eligible for these funds, however, meaning the connection between community health organizations (like those in behavioral health) often have to finance the purchase of these systems independently.While the information and data collected through EHR can be utilized in ways that promote health, reduce redundancies, and make information more readily available to providers and patients, they also present a number of challenges. On one level, each are the very practical considerations for organizations who want to use this technology. Organizations have to have the capacity, training protocol and ongoing support in place to have these systems available and functional. Doctors, for example, can benefit from automated prompts that remind them of evidence-based practices, but may also end up spending less time speaking with patients in the normal course of care delivery while navigating a complex software system. On another level, while ODM and MCOs are able to access much of the outcome data, they do not have direct access to the records themselves, meaning there may be gaps in understanding population health trends and individual patient needs.There are also broader questions of security and interoperability. On the security front, there has been an uptick in stories about hospital systems having to pay ransoms to hackers who have successfully captured patient data. [19]  This phenomenon of hacking is in addition to existing provider anxiety about the best way to exchange data while remaining compliant with the Health Information and Health Insurance Portability and Accountability Act of 1996 (HIPPA). While HIPPA is meant to simultaneously protect and promote the exchange of data, it is often cited as a barrier in sharing useful information between provider groups. Even when this data can be shared, the issue of interoperability can be a concern. With no standard, mandatory EHR product to be used by all providers, many systems implement those record systems which meet their unique needs, often creating difficulties in the exchange process. This is why states and providers often invest in the creation of Health Information Exchanges (HIE).Ohio’s largest HIE is the Ohio Health Information Partnership (OHIP), also known as CliniSync. CliniSync is a nonprofit organization that serves a centralizing role in creating interoperability between record systems. Depending on participation and the nature of the activity in question, HIE can overcome the barriers created by differentiation through the standardization of clinical information. The HIE also serves a security role for providers sending and retrieving patient data by using systems that encrypt shared data. This sharing, according to CMS, can be direct (between providers), query-based (request-based) or consumer-mediated (the patient directs the nature of the exchange).[20] While the use of OHIP is not mandatory, HIE’s role in navigating disparate record-keeping systems is a model that will be ever-present. Whether it’s within a third party entity (like OHIP), vertically integrated (like a provider with an insurance product) or horizontally integrated (like a trade association with a common contract for its members), the exchange of health data will continue to be a need in the delivery system. VALUE-BASED DESIGNAccording to recent research from Stanford Medicine, data in health care is growing 48 percent a year.[21] Data in health care, however, is not limited to the influence or preferences of Medicaid programs. To be sure, there are a number of spaces and applications where this data will be used, for better or for worse, depending on perspective. Included in this realm of possibilities is refining the patient experience, utilization of wearables in the monitoring of health conditions (such as high blood pressure) and development of advanced analytics programs which will be able to predict certain health outcomes before an intervention is needed. A recent Manatt report highlights many of these trends in the intersection of data and health care and has identified a potential role of government in the standardization of the mechanics of exchange as well as the regulatory environment in the interest of greater interoperability and greater sharing.[22]OHT has invested in a number of program integrity efforts, including information technology infrastructure, analytics and electronically based systems of billing for providers (such as the electronic visit verification system for home health providers).[23]  While on the surface, these sorts of developments seem perfunctory or, at least, secondary to the state’s overall policy efforts with Medicaid, they are anything but.In its initial documentation concerning its policy vision, OHT characterized Ohio’s delivery system as fragmented and low quality. While it is often the policy decisions around Medicaid expansion and waivers that get the most attention, politically, the greatest long-term effects come from the investments in reorganization of the governance behind the program, as well as the tools implemented to centralize and digitize the benefit administration process. Put in more simplistic terms, the money in Medicaid is now more measurable than it has ever been. Providers are constantly feeding in health information that is tied to other pieces of data, including eligibility, demographics, provider-type, price, effectiveness and outcomes.

...the money in Medicaid is now more measurable than it has ever been.

The episodic-based payment program, for example, is ultimately a value-based data project. Providers are reporting common billing codes of service delivery which are then used to create comparisons between peers on quality and price. The state, through its policy-making, creates incentives that put more weight behind those activities they feel produce the highest value. To do this however, the state needed to have standardized provider enrollment processes, uniform contracting between the plans and the state, common billing infrastructure and standardized eligibility. In other words, without data, and the infrastructure to support it, the value-based work of the state in Medicaid would not be as achievable.CONCLUSIONAs the largest single purchaser of health care benefits, the State of Ohio’s Medicaid program faces unique market forces that compel innovation. First, one of the most expensive categories of beneficiaries--the elderly--is the fastest growing demographic in the state. Second, Ohio’s second largest industry -- health care -- is fundamentally dependent on the policy decisions of lawmakers and the Ohio Department of Medicaid to be made whole and exist as a provider and a tool in the state’s fight against some of the most pressing public health crises like the opioid epidemic. Third, despite having some of the most remarkable medical institutions in the world, Ohio is still lagging the nation in its health rankings – a dubious recognition given the United States’ position in lagging other industrialized nations for the same reasons.[24][25]Medicaid has had a historically simple role as a publicly subsidized insurance program for the aged, disabled and those in poverty, particularly children. As Medicaid has grown in size and complexity, there has been greater pressure to control spending and seek ways to rationalize the program, despite the marked history of poor outcomes. Politically, the tools for these cost control efforts have been blunt and simplistic: reduce benefits, limit enrollment, and require more of consumers. Emphasis has also been placed, appropriately, on controlling fraud, waste and abuse through extensive multilateral partnerships to prevent illegal activities. At the end of the day, however, the cost in the system stems not from the inappropriate actions of consumers or providers, but rather from the inefficient process of purchasing that defines the American health care delivery system.Data carries the potential to enact value-based reform and to be used as a tool of politics. It is clear that the Kasich administration is able to use data in both ways, with significant efforts to build the infrastructure that enables the use of data in eligibility, claims, auditing and contracting as well as implementing a political agenda. While Ohio’s overall value is still lagging behind other states, this effort has enabled some of the most promising work in value-based design, including Ohio’s efforts around episodic-based payments, pay for performance and the comprehensive primary care program. It has also been used to highlight the need to fundamentally change the program through privatization, benefit expansions and rate reforms. It remains to be seen how either of these types of efforts will continue regardless of who is the next governor of Ohio, but the potential to build upon the policy utility of these data efforts cannot be overlooked and should create common cause between policymakers, regardless of ideological position.[1] "Program Integrity Report 2015." 2015. Accessed June 15, 2018. http://www.jmoc.state.oh.us/Assets/documents/reports/program-integrity-2015.pdf[2] "Simplify and Integrate Eligibility Determination." Accessed June 3, 2018. http://healthtransformation.ohio.gov/Current-Initiatives/Modernize-Eligibility-Determination-Systems.[3] "Performance Indicator Technical Assistance." Accessed June 6, 2018. https://www.medicaid.gov/medicaid/program-information/medicaid-and-chip-enrollment-data/sdis/index.html.[4] Cahill, Rachel. "Ohio Benefits Pilot Update." The Center for Community Solutions. April 19, 2018. Accessed June 22, 2018. https://www.communitysolutions.com/ohio-benefits-pilot-update/.[5] Anthes, Loren. "Work Requirement Waiver." The Center for Community Solutions. February 21, 2018. Accessed June 22, 2018. https://www.communitysolutions.com/work-requirement-waiver/.[6] "Group VIII Work Requirement and Community Engagement 1115 Demonstration Waiver Application." The Ohio Department of Medicaid. April 30, 2018. Accessed May 26, 2018. http://medicaid.ohio.gov/Portals/0/Resources/PublicNotices/GroupVIII/GroupVIII-WRCE-1115-Final-Submission.pdf.[7] Edes, Alyssa, and Emma Bowman. "'Automating Inequality': Algorithms In Public Services Often Fail The Most Vulnerable." NPR. February 19, 2018. Accessed June 22, 2018. https://www.npr.org/sections/alltechconsidered/2018/02/19/586387119/automating-inequality-algorithms-in-public-services-often-fail-the-most-vulnerab.[8] “Federal Medical Assistance Percentage (FMAP) for Medicaid and Multiplier.” The Henry J Kaiser Family Foundation. Accessed June 2, 2018. https://www.kff.org/medicaid/state-indicator/federal-matching-rate-and-multiplier/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D.[9] "Implement a New Medicaid Claims Payment System." Ohio Office of Health Transformation. Accessed June 6, 2018. http://www.healthtransformation.ohio.gov/Current-Initiatives/Medicaid-Claims-Payment-System[10] "Ohio MITS for Providers." Coleman Data Solutions. January 22, 2014. Accessed June 22, 2018. http://www.coleman-data.com/paperplus/ohio-mits/.[11] "Program Integrity." The Centers for Medicare and Medicaid Services. Accessed June 22, 2018. https://www.medicaid.gov/medicaid/program-integrity/index.html.[12] The Ohio Department of Medicaid, Program Integrity Report 2015[13] U.S. Government Accountability Office. "Medicaid Program Integrity: CMS Should Build on Current Oversight Efforts by Further Enhancing Collaboration with States." U.S. Government Accountability Office (U.S. GAO). April 17, 2017. Accessed June 14, 2018. https://www.gao.gov/products/GAO-17-277.[14] Erlewine, Kristi. "Medicaid Program Integrity A Detailed Perspective." Speech. Accessed May 18, 2018. https://ohioauditor.gov/trainings/miaf/TrainingDocs/Day 2/Medicaid Program Integrity.pdf.[15] "2017 Ohio Medicaid Fraud Control Unit Annual Report." Ohio Attorney General. 2017. Accessed June 22, 2018. http://www.ohioattorneygeneral.gov/getattachment/bc04f0c2-0b1a-4142-bf4c-f404641331c0/2017-Health-Care-Fraud-Annual-Report.aspx.[16] Smith, Vernon K., Kathleen Gifford, Eileen Ellis, Robin Rudowitz, and Laura Snyder. "Medicaid in an Era of Health & Delivery System Reform: Results from a 50-State Medicaid Budget Survey for State Fiscal Years 2014 and 2015 - Program Integrity Initiatives." The Henry J. Kaiser Family Foundation. April 14, 2016. Accessed June 7, 2018. https://www.kff.org/report-section/medicaid-in-an-era-of-health-delivery-system-reform-program-integrity-initiatives/.[17] "Office-based Physician Electronic Health Record Adoption." Dashboard.healthit.gov. Accessed June 22, 2018. https://dashboard.healthit.gov/quickstats/pages/physician-ehr-adoption-trends.php.[18] "Medicare & Medicaid EHR Incentive Program." The Centers for Medicare and Medicaid Services. 2010. Accessed May 21, 2018. https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/downloads/MU_Stage1_ReqOverview.pdf.[19] Ryckaert, Vic. "Hackers Held Patient Data Ransom, so Indiana Hospital System Paid $50,000." USA Today. January 17, 2018. Accessed June 18, 2018. https://www.usatoday.com/story/tech/nation-now/2018/01/17/hackers-held-patient-data-ransom-so-indiana-hospital-system-paid-50-000/1042266001/.[20] "What Is HIE?" HealthIT.gov. Accessed June 22, 2018. https://www.healthit.gov/topic/health-it-and-health-information-exchange-basics/what-hie.[21] "Stanford Medicine 2017 Health Trends Report: Harnessing the Power of Data in Health." Standford Medicine. 2017. Accessed June 10, 2018. https://med.stanford.edu/content/dam/sm/sm-news/documents/StanfordMedicineHealthTrendsWhitePaper2017.pdf.[22] Barton, Valarie, Laura Braslow, and Kevin Casey McAvey. "The Promise of Data-Driven Healthcare: MegaTrends for 2018 and Beyond." Manatt Phelps Phillips. February 27, 2018. Accessed June 22, 2018. http://img.en25.com/Web/ManattPhelpsPhillipsLLP/{4b1a5ad6-9f9c-4bdf-b6e2-1e0ae62e7f3f}_Data_MegaTrends_Presentation_20180227_0830_Final_v2.pdf.[23] "Fight Medicaid Fraud and Abuse." Ohio Office of Health Transformation. February 2, 2015. Accessed June 16, 2018. http://healthtransformation.ohio.gov/Portals/0/1H Fraud.pdf?ver=2015-02-02-141810-220.[24] Viviano, JoAnne. "Ohio Still Lags Nation in Health Rankings." The Columbus Dispatch. January 07, 2018. Accessed June 22, 2018. http://www.dispatch.com/news/20180107/ohio-still-lags-nation-in-health-rankings.[25] Osborn, Robin, David Squires, Michelle M. Doty, Dana O. Sarnak, and Eric C. Schneider, MD. "In New Survey of 11 Countries, U.S. Adults Still Struggle with Access to and Affordability of Health Care." The Commonwealth Fund. November 16, 2016. Accessed June 2, 2018. https://www.commonwealthfund.org/publications/journal-article/2016/nov/new-survey-11-countries-us-adults-still-struggle-access-and.