Legal Reforms Necessary to Promote Delivery System Innovation

At the center of any serious health care reform must be reform of the delivery system. The current fragmented fee-for-service system produces haphazard quality, fosters the use of unproven interventions, and increases costs. Although this delivery system fails to generate optimal results, the best alternative is unknown. Indeed, it is unlikely that there is a single best way to organize and deliver health care services. Sustainable reform must facilitate innovation by fostering different ways of organizing and paying for health care services.¹

The current legal environment has created major barriers to delivery system innovation. Innovation will not occur if each novel way to organize and pay for care needs to be adjudicated case-by-case or is threatened with legal proceedings. What are the problems with the current delivery system? What legal reforms are needed to encourage delivery system innovation?

The current health care system is too fragmented. Patients in the United States make approximately 1 billion office visits per year—nearly 40% to solo practitioners and 37% to physicians working in groups of 2 to 5.² Fragmentation often results in uncoordinated care. Each year, Medicare patients see an average of 7 different physicians, but these physicians are usually not linked clinically, electronically, financially, or administratively.³

An increasing body of data suggests that care received from multiple, uncoordinated physicians is often of poor quality.^{4 - 5} Conversely, patients who receive care from integrated medical groups receive on average higher-quality medical care and are more likely to receive proven effective interventions.⁶

Fragmentation has many sources: historical, cultural, financial, legal, and others. For instance, many physicians are individualists who want autonomy over their work life. The fee-for-service reimbursement system inhibits coordination of care between physicians, hospitals, visiting nurses, and other health care personnel and facilities.⁷

To improve the quality of care and restrain costs require coordinated care, including a reimbursement system that facilitates such coordination. However, the optimal way to organize care and pay clinicians is unknown. A variety of alternatives are being tried. For instance, Shortell and Casalino⁸ have described 5 different models of “accountable care systems” with varying degrees of clinical, administrative, and financial integration. Similarly, trials of pay-for-performance, bundled payments, partial capitation, and other payment systems are ongoing. The United States needs a continuous process of experimentation with organizational and payment innovations to provide high-quality care at appropriate costs.

One of the biggest barriers to delivery system innovation is the complex web of laws and regulations. Multiple and conflicting laws at both the federal and state levels, the legal morass that governs Medicare, public and private regulatory programs, such as The Joint Commission accreditation of hospitals, all inhibit organizational experimentation and delivery system innovation (Table).

In many states, the corporate practice of medicine doctrine still prohibits corporations other than professional corporations from hiring health care professionals.⁹ For instance, in California this doctrine would seem to require that the medical staff of store-based clinics be employed by medical corporations, not by retail stores. However, the doctrine probably no longer applies in many jurisdictions, but in others its status is unclear, while in yet others it persists and must be evaded by organizational structures that confound incentives for coordination. States' restrictive scope of practice limitations impede the ability of health care delivery systems to optimize the use of their professionals, such as nurse practitioners. In more than two-thirds of the states, certificate of need laws contribute to the difficulty of establishing innovative health care facilities.¹⁰

Federal laws, such as provisions in the Department of Justice/Federal Trade Commission Statements of Antitrust Enforcement Policy in Health Care, permit some physician joint ventures on the basis of financial risk sharing and permit others on the basis of clinical integration.¹¹ The type and level of integration sufficient to shield an arrangement from antitrust scrutiny are difficult to predict; this uncertainty could deter attempts to create accountable health systems.

Even more problematic are federal antikickback and self-referral prohibitions. The antikickback statute criminalizes the knowing and willful offer or acceptance of remuneration for referrals of patients financed by federal health care programs.¹² The Stark II self-referral statute prohibits physicians from referring Medicare or Medicaid patients for designated health care services where the physician has an ownership or compensation relationship with the entity to which the patient is referred.¹³ These statutes affect almost all financial relationships between physicians and other medical facilities, such as free-standing surgery centers or visiting nurse companies. A host of exceptions and “safe harbors” explicitly permit some relationships, while “fraud alerts” attempt to deter others, but many possible arrangements are left in a state of uncertainty. This uncertainty is compounded by hundreds of pages of administrative regulations and commentary. The 2 prohibitions are largely redundant but contain different exceptions, making compliance more problematic.

Tax laws limit joint ventures between tax-exempt and for-profit organizations, while also constraining the benefits that exempt organizations can offer professionals. The federal prohibition against hospitals offering physicians inducements to reduce services to Medicare beneficiaries has mitigated “gain-sharing” arrangements between hospitals and physicians that reward the prudent use of services.¹⁴ In addition, the Medicare statute divides the health care system into “silos,” with separate prospective payment systems, participation requirements, and even cost-sharing arrangements for different types of physicians and other medical facilities.

The wholesale repeal of these laws is neither possible nor advisable. They enshrine important principles: competitors should not conspire to undermine competition; referrals should be based on the patient's medical needs, not the physician's financial interest. Rationalization—not abdication—is necessary.

Some laws should indeed be eliminated. Many commentators have long called for the abolition of the corporate practice of medicine doctrine. Scope of practice laws have been effectively defined in more functional terms in some states and countries. The compensation portion of the self-referral statute largely duplicates the bribe and kickback prohibition but greatly complicates transactional analysis because it has its own elaborate set of exceptions and special rules.

Repealing laws on a state-by-state basis would be difficult, but some state laws could be overridden through a general federal law, as the Employee Retirement Income Security Act does for much of state insurance regulation. Others could be preempted insofar as they affect federally funded programs.

Another approach would be more appropriate for most federal laws. Currently, several federal agencies are permitted to authorize health care arrangements subject to their jurisdiction that may violate the law. The US Department of Health and Human Services Office of Inspector General can offer advisory opinions in applying the antikickback law. The Internal Revenue Service issues revenue rulings or private letter rulings addressing exempt organizational arrangements. The Federal Trade Commission issues advisory opinions interpreting the antitrust laws. The Centers for Medicare & Medicaid Services offers advisory opinions regarding the self-referral laws and approves Medicare and Medicaid demonstration projects. Currently, proposals for health care delivery innovation often require knocking on many doors.

An alternative is to create a single governmental Commission for Innovation in Delivery Systems (CIDS) with representatives from each of these agencies, plus the Department of Justice. The CIDS would offer one-stop review to permit experimentation in delivery systems. With statutory authorization, advisory opinions from the CIDS could go further than those currently allowed, authorizing not only proposals that pose no threat of violation of the law, but also arrangements that are consistent with the goals of federal statutes. Legislation could further preempt the enforcement of certain state laws against approved entities, as Congress preempted state laws restricting federally qualified health maintenance organizations in the 1970s.¹⁵ The CIDS could be charged with authorizing innovative approaches to Medicare payment for proposed delivery arrangements as Medicare demonstration projects. The CIDS, for example, could approve Medicare payment for a new coordinated but not fully integrated accountable care system that otherwise might raise questions under various laws.

As the current banking crisis illustrates, innovation may not always be progress. Several safeguards can minimize potential risks. The CIDS approvals could have a sunset provision with renewals conditionally linked to formal evaluation by the Agency for Healthcare Research and Quality or the Medicare Payment Advisory Commission (MedPAC) confirming that the innovations have lowered costs, improved quality, or both. Effective review and oversight would require adequate resources and staffing. To provide such funds—and minimize frivolous submissions—a fee calibrated to the number of physicians and other clinicians or beneficiaries affected might be assessed on proposals submitted to CIDS. The entire process would need to be transparent to minimize serious administrative abuses.

New arrangements that proved successful could become models for new safe harbors under the tax, antitrust, antikickback, and self-referral laws and as templates for similar proposals, which could in turn be granted accelerated approval by the CIDS. The success of certain arrangements could lead to proposals for amending laws that deter wider diffusion.

The current health care delivery system is dysfunctional, is excessively costly, and provides haphazard quality. Although not without risk, significant reform is necessary because maintaining the status quo is untenable. A host of federal and state laws now complicate innovation within the delivery system. A federal commission with authority to permit delivery system innovations that are time limited and contingent on periodic evaluations demonstrating cost savings and improved quality could facilitate the necessary reforms.